Free Will and God’s Universal Causality: The Dual Sources Account 9781350082908, 9781350082939, 9781350082915

Citation preview

Free Will and God’s Universal Causality

Bloomsbury Studies in Philosophy of Religion Series Editor: Stewart Goetz Editorial Board: Thomas Flint, Robert Koons, Alexander Pruss, Charles Taliaferro, Roger Trigg, David Widerker, Mark Wynn Titles in the Series Freedom, Teleology, and Evil by Stewart Goetz The Image in Mind: Theism, Naturalism, and the Imagination by Charles Taliaferro and Jil Evans Actuality, Possibility, and Worlds by Alexander Robert Pruss The Rainbow of Experiences, Critical Trust, and God by Kai-man Kwan Philosophy and the Christian Worldview: Analysis, Assessment and Development edited by David Werther and Mark D. Linville Goodness, God and Evil by David E. Alexander Well-Being and Theism: Linking Ethics to God by William A. Lauinger Thinking through Feeling: God, Emotion and Passibility by Anastasia Philippa Scrutton God’s Final Victory: A Comparative Philosophical Case for Universalism by John Kronen and Eric Reitan Free Will in Philosophical Theology by Kevin Timpe Beyond the Control of God? edited by Paul M. Gould The Mechanics of Divine Foreknowledge and Providence by T. Ryan Byerly The Kalām Cosmological Argument: Philosophical Arguments for the Finitude of the Past edited by Paul Copan with William Lane Craig The Kalām Cosmological Argument: Scientific Evidence for the Beginning of the Universe edited by Paul Copan with William Lane Craig

Free Will and God’s Universal Causality The Dual Sources Account W. Matthews Grant

BLOOMSBURY ACADEMIC Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA BLOOMSBURY, BLOOMSBURY ACADEMIC and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2019 Paperback edition first published in 2020 Copyright © W. Matthews Grant, 2019 W. Matthews Grant has asserted his right under the Copyright, Designs and Patents Act, 1988, to be identified as Author of this work. For legal purposes the Acknowledgments on p. vii constitute an extension of this copyright page. Series design by Louise Dugdale Cover image © Daniel Bosma / Getty Images All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Grant, W. Matthews, author. Title: Free will and God's universal causality: the dual sources account / W. Matthews Grant. Description: 1 [edition]. | New York: Bloomsbury Academic, 2019. | Series: Bloomsbury studies in philosophy of religion | Includes bibliographical references and index. Identifiers: LCCN 2018050930 (print) | LCCN 2019011249 (ebook) | ISBN 9781350082915 (epdf) | ISBN 9781350082922 (epub) | ISBN 9781350082908 (hardback) Subjects: LCSH: Free will and determinism. | Causation. | Theism. | Teleology. | Free will and determinism–Religious aspects–Christianity. Classification: LCC BJ1461 (ebook) | LCC BJ1461 .G73 2019 (print) | DDC 123/.5–dc23 LC record available at https://lccn.loc.gov/2018050930 ISBN: HB: 978-1-3500-8290-8 PB: 978-1-3502-0365-5 ePDF: 978-1-3500-8291-5 eBook: 978-1-3500-8292-2 Series: Bloomsbury Studies in Philosophy of Religion Typeset by Integra Software Services Pvt. Ltd., To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

Contents Acknowledgments 1 Introduction 1.1 Divine Universal Causality (DUC) and Creaturely Action 1.2 Libertarian Freedom and the Apparent Conflict with DUC 1.3 Dual Sources: A Neo-scholastic Approach to Resolving the Conflict

vii 1 1 5 10

2 God: Universal Cause and Cause of Human Actions 2.1 Scripture 2.2 Perfect Being Theology: An Anselmian Approach 2.3 Cosmological Arguments from Contingency 2.4 Conservation and Concurrence: A Suarezian Argument 2.5 A Thomistic Argument from Participation

15

3 Divine Universal Causality and the Threat of Occasionalism 3.1 Does DUC Render Creaturely Causes Otiose? 3.2 God and Creaturely Causes: The Claims of Non-Occasionalist DUC 3.3 Non-Occasionalist DUC: The Metaphysical Objection 3.4 Non-Occasionalist DUC: The Epistemic Objection 3.5 Can Agent-causal Acts Be Caused by God?

35

4

5

15 22 24 25 29

35 37 41 45 48

Free Creatures of the Universal Cause 4.1 The Intrinsic/Extrinsic Distinction 4.2 Why DUC May Appear to Preclude Libertarian Freedom 4.3 The Extrinsic Model of Divine Agency 4.4 DUC without Determinism 4.5 Ability to Do Otherwise 4.6 Ultimate Responsibility 4.7 Dual Sources

53

The Extrinsic Model Defended 5.1 The Extrinsic Model, Intrinsic Models, and Scholastic Theology 5.2 From DUC to the Extrinsic Model 5.3 But Is the Extrinsic Model Also Ruled Out by DUC? 5.4 Does the Extrinsic Model Render Divine Causality Unintelligible?

75

53 54 56 60 65 68 70

75 80 87 92

Contents

vi 6

7

8

Does God Cause Sin? 6.1 DUC, Moral Evil, and the Privation Solution 6.2 Moral Evil and Privation 6.3 Objections to the Privation Account of Moral Evil 6.4 Does God Cause the Badness in Sinful Acts Simply by Causing the Acts? 6.5 How the Badness in Sinful Acts Is Caused by the Sinner Alone

99 99 103 108 112 114

The Problem of Moral Evil 7.1 The Failure of the Free Will Defense 7.2 Responding to the Problem without FWD 7.3 Moral Evil, Dual Sources, and Molinism 7.4 Moral Evil, Dual Sources, and Open Theism 7.5 Sin and the Divine Will 7.6 God’s Involvement in Sin: A Cost of Dual Sources?

119

Providence, Grace, and Predestination 8.1 An Extrinsic Model of Divine Knowing 8.2 Time, Foreknowledge, and a Variation on the Eternity Solution 8.3 Providence 8.4 Grace 8.5 Divine–Human Dialogue 8.6 Predestination

145

Notes Bibliography Index

119 124 126 128 134 140

145 150 155 158 166 171 182 231 243

Acknowledgments I would like to thank my undergraduate professors at Wake Forest University, especially Jeffery Kinlaw, Charles Lewis, and Ralph C. Wood; the existence of this book testifies to their enduring influence. I would like to thank my graduate professors at Fordham University, in particular Brian Davies, John Greco, and Joseph Koterski, who served as the committee for my dissertation where ideas for the book took incipient form almost two decades ago. I would like to thank my colleagues from the department of philosophy at the University of St. Thomas; I am a much better and happier philosopher for having such wonderful examples, teachers, and friends as you have been. I would like to thank the University of St. Thomas for support through a sabbatical and a research grant, and Thomas D. Sullivan for support through his Aquinas Chair. I would like to thank Alfred Mele, the John Templeton Foundation, and the Big Questions in Free Will Project out of Florida State University for generously funding a yearlong Theology of Free Will fellowship for work on this project in 2012–2013. Although most of this book is new, I am grateful for permission to reuse or draw from some previously published material. Chapter 3 is taken largely from my “Divine Universal Causality without Occasionalism (and with Agent-Causation),” published in Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology (Oxford: Oxford University Press, 2017), edited by Hugh J. McCann. Much of Chapter 4, though it includes important developments, draws substantially from my “Can a Libertarian Hold That Our Free Acts Are Caused by God,” Faith and Philosophy 27 (2010): 22–44, and my “Divine Universal Causality and Libertarian Freedom,” published in Free Will & Theism: Connections, Contingencies, and Concerns (Oxford: Oxford University Press, 2016), edited by Kevin Timpe and Daniel Speak. Chapter 6 is taken largely from my “The Privation Account of Moral Evil: A Defense,” International Philosophical Quarterly 55 (2015): 271–86, and my “Moral Evil, Privation, and God,” European Journal for Philosophy of Religion 9 (2017): 125–45. Finally, a small portion of Chapter 8 draws from my “Divine Simplicity, Contingent Truths, and Extrinsic Models of Divine Knowing,” Faith and Philosophy 29 (2012): 254–74. I owe special gratitude to those who have offered written comments on various parts of the book and those who have helped me by extensive correspondence or conversation on its topics. My colleague Tim Pawl deserves first mention for commenting on one or another version of every chapter, serving as a frequent sounding board for arguments and ideas, and being a continual source of encouragement. Others who have helped with this project, in some cases immensely, by way of comments, conversation, or encouragement include Gary M. Atkinson, Stephen L. Brock, Jeffrey E. Brower, David. B. Burrell, David Clemenson, Brian Davies, Nicolas Feddema, Thomas P. Flint, Gloria Frost, the late Germain Grisez, Jonathan D. Jacobs, Andrew Jaeger, Neal Judisch, Joseph Koterski, John D. Kronen, Patrick Lee, Stephen A. Long, James D. Madden,

viii

Acknowledgments

R.J. Matava, the late Hugh J. McCann, Ross McCullough, Sandra L. Menssen, Chris Mullen, Thomas M. Osborne Jr., Alexander R. Pruss, Katherin A. Rogers, Michael Rota, Mark K. Spencer, Jaime Spiering, Eleonore Stump, Kevin Timpe, Michael Torre, and my Philosophy of God students over the years. My sincere apologies to anyone I have missed. And my gratitude, also, to those I reference in the book, from whom I have learned much. I would like to thank my editors at Bloomsbury, especially Colleen Coalter, Becky Holland, and Helen Saunders, for all they have done to bring this book to print. I would also like to thank the series editor, Stewart Goetz, as well as the anonymous reviewers of the proposal and manuscript. Finally, I am deeply grateful to my dear wife, Kristen, and our daughters, Elizabeth and Cecilia, for their love, support, and the sacrifice of much family time so that I could see this project through. To these I add my loving and supportive parents, Walter and Ann Grant. All of these have given me more than I can say and certainly more than I deserve. I dedicate this book to my parents, Kristen, and the girls.

1

Introduction

According to traditional theism, God is the universal cause, who causes all being distinct from himself. An implication of this doctrine is that creaturely acts are caused by God. With few exceptions, contemporary philosophers assume that a creaturely act caused by God cannot be free in the libertarian sense. The present volume challenges this assumption, showing that libertarian theists need not reject a central tenet of the classic theological tradition. In the course of this challenge, a comprehensive alternative to the currently most popular approaches for combining theism with libertarian freedom will emerge. In this introductory chapter, I first set out the doctrine of divine universal causality (DUC), displaying its deep roots within theistic tradition, giving the doctrine a precise definition, and drawing out its implications for God’s relationship to creaturely action. Then, after presenting the libertarian conception of freedom and explaining why it has proven attractive to many theists, I give evidence of, and suggest reasons for, the widespread assumption that a creaturely act being caused by God precludes the act being free in the libertarian sense. Finally, I introduce my approach to resolving the conflict—what I call “Dual Sources”—offering a brief preview of how the approach will be developed in the chapters that follow and explaining the approach’s roots in the scholastic philosophical-theological tradition.

1.1 Divine Universal Causality (DUC) and Creaturely Action Theists have traditionally held that God is the universal cause, who causes all being distinct from himself. Augustine, for example, teaches that God’s “hidden power … causes all that exists in any way to have whatever degree of being it has; for without Him, it would not exist in this way or that, nor would it have any being at all.”1 Anselm maintains that “with the exception of the supreme essence itself [God], nothing exists that is not made by the supreme essence.”2 Maimonides writes that “the basic principle of all principles and the pillar of all sciences is to realize that there is a First Being who brought every existing thing into being.”3 Aquinas holds that “everything other than God … must be referred to Him as the cause of its being.”4 Such belief in God’s universal causality is rooted in scripture, which declares God maker of all things (Isa. 44:24) and the one from whom and through whom all things are (Rom. 11:36).5 The

2

Free Will and God’s Universal Causality: The Dual Sources Account

Church Fathers strongly affirm God’s universal causality.6 The Nicene Creed expresses such belief by confessing God “maker of heaven and earth, of all things visible and invisible.”7 With such impressive support from the tradition, it is not surprising that contemporary philosophers of religion often include God’s universal causality in accounts of what theists have traditionally believed. Thus, Alvin Plantinga notes that most Christians affirm God’s “control over all things and the dependence of all else on his creative and sustaining activity.”8 According to Brian Leftow, “theists believe that necessarily, for any x, if x is God, x creates and maintains in existence whatever is not identical to x.”9 And Thomas Morris remarks that “arguably the central idea of the theistic tradition” is “the idea of God as absolute creator of everything which exists distinct from him.”10 Statements of this sort are easily multiplied. Michael Bergmann and Jeffrey Brower consider “traditional theism” to hold that “everything distinct from God depends on God’s creative activity for its existing.”11 Hugh McCann explains that “the concept of a creator as it has been usually construed in the Western theological tradition” involves “the idea that the world and all that pertains to it—indeed, anything that exists in any way—owes its being and sustenance to the act of an all-powerful being whose existence requires no explanation.”12 William Alston concurs that “God is the ultimate source of being for everything other than himself.”13 And Katherin Rogers characterizes “traditional, classical theism” as holding that “whatever has any positive ontological status at all is God or comes from God.”14 What, then, is numbered among the effects God causes? Some of the foregoing statements are more explicit than others, but the natural reading of most of them would suggest that God causes every entity whatsoever that is not identical to God. Whether this reading is the one intended by all the authors quoted, it is certainly the traditional view. It is not, therefore, only creaturely objects or substances that God has traditionally been thought to cause. As Anselm insists, “every quality, every action, everything that has existence owes its being at all to God.”15 As Aquinas makes clear, “whatever is the cause of things considered as beings [namely, God], must be the cause of things, not only according as they are such by accidental forms, nor according as they are these by substantial forms, but also according to all that belongs to their being at all in any way.”16 As some of the passages quoted above suggest, the traditional view understands the being of all that is distinct from God to be caused by God, not just at the first moment such things come into existence but rather for the entire duration of their existence. Aquinas is representative: Since God is very being by His own essence, created being must be His proper effect; … Now God causes this effect in things not only when they first begin to be, but as long as they are preserved in being; as light is caused in the air by the sun as long as the air remains illuminated.17

The belief that God keeps things in being is usually expressed by saying that God “preserves,” “sustains,” or “conserves” them; but it would be a mistake to read the

Introduction

3

tradition as claiming that a thing’s being “conserved” by God denotes an essentially different sort of dependence on God than its being caused to exist at its first moment. Jonathan Kvanvig and Hugh McCann helpfully explain the traditional view as one according to which each instant of the existence of any of God’s creatures is as radically contingent as any other, and equally in need of activity on His part to account for it. Furthermore, God’s activity as Creator is held to be essentially the same no matter what instant of the being of the thing created is at stake.18

God’s causality, then, has traditionally been thought to extend to his effects in the same way, at any time they exist. Moreover, the traditional view maintains not just that everything distinct from God is, in fact, caused by God but that it is not possible that anything else exist without being caused by him. Thus, Anselm, having concluded that the supreme essence makes everything other than itself, goes on to argue that “something else is only able to exist where and when the supreme essence exists … without it absolutely nothing exists.”19 Maimonides likewise holds that “if it could be supposed that He [God] did not exist, it would follow that nothing else could possibly exist.”20 Aquinas teaches that “there can be nothing besides [God] that is not caused by Him.”21 And Plantinga remarks that “it is not possible, at least if traditional theism is correct, that we should exist and God not create and sustain us.”22 Finally, the traditional view understands God’s causing of things distinct from himself to be immediate and direct, rather than mediate, indirect, or remote. So, Aquinas, in arguing for God’s omnipresence, states that “God is in all things … as an agent is present to that upon which it works. For an agent must be joined to that wherein it acts immediately.”23 Here, Aquinas is offering God’s immediate causal action in all things distinct from himself as a premise in an argument for God’s omnipresence. To the objection that God’s power is such that his action even extends to things distant from him, Aquinas responds: No action of an agent, however powerful it may be, acts at a distance, except through a medium. But it belongs to the great power of God that He acts immediately in all things.24

Like Aquinas, Suarez maintains that God causes his effects directly and immediately,25 describing divine conservation, for instance, as involving God’s “direct and immediate continuous inpouring of the esse itself within one and the same singular entity.”26 Among contemporary philosophers, Morris is especially clear on the direct nature of the dependence that all things other than God have on God: All things distinct from God stand in a dependence relation to God, a relation that is both direct and absolute. It is never the case that some created object x depends on God only in the sense of depending for its existence upon some other created objects y and z, which in turn directly depend on God. … Metaphysical or ontological dependence upon God, dependence for being, is, rather, in every case direct.27

4

Free Will and God’s Universal Causality: The Dual Sources Account

To say that God causes his effects directly or immediately, I take it, is to deny that he causes any entity a by means of causing some other entity b, which is a more proximate cause of a or which is that by means of which, or in virtue of which, God causes a. On the traditional view, then, every divine causing of an effect is a “basic” causal action, a causal act God performs, but not by means of causing anything else.28 Incorporating the various elements of the traditional view, we can now define the doctrine of divine universal causality (DUC) as follows: Necessarily, for any entity distinct from God, God directly causes that entity to exist at any time it exists.29

Notice, DUC does not claim that God is the only genuine cause, as in occasionalism.30 It does imply, however, that if there are other causes, it is not possible that those causes bring about their effects without God’s also directly bringing those effects about. Moreover, it implies that God directly causes all creaturely actions, whether free or unfree, since whatever they consist in, creaturely actions are entities distinct from God.31 Thus, Brian Davies notes: Traditionally speaking, all things apart from God are there because God makes them to be there, not just in the sense that he lays down the conditions in which they can arise, but also in the sense that he makes them to be for as long as they are there. And on this account, all that is real in creatures is caused by God, including their activity.32

This implication for God’s relationship to creaturely action was unhesitatingly acknowledged by classical proponents of DUC. Anselm’s “student” no doubt expresses Anselm’s own conclusion on the matter: “I cannot in fact deny that every action is a reality nor that whatever is has its being from God.”33 Aquinas holds that God “acts in every agent immediately”;34 that “we must admit without any qualification that God operates in the operations of nature and will”;35 and that “the very act of free will is traced to God as to a cause.”36 Suarez affirms that “God acts per se and immediately in every action of a creature, and that this influence of his is absolutely necessary in order for the creature to effect anything.”37 Consistency requires, of course, that the proponent of DUC admit that even evil or sinful acts are caused by God, an implication not lost on the doctrine’s traditional proponents. Thus, according to Aquinas, “the act of sin is both a being and an act; and in both respects it is from God. Because every being, whatever the mode of its being, must be derived from the First Being, as Dionysius declares.”38 Similarly, Suarez holds that “one should deny that any action (be it natural or free, good or evil), insofar as it is a real action, exists without the First Cause’s immediate concurrence.”39 As illustrated in this last passage from Suarez, the term “concurrence” came to be used by scholastics to refer to God’s causal activity in co-operating to bring about the effects and actions of creaturely agents when they act. Scholastics generally affirm that, in addition to God the universal cause, there are creaturely or secondary causes that

Introduction

5

exercise genuine causal powers in bringing about effects in the world. Given DUC, however, it is not possible for any creaturely agent to act or to bring about an effect without God’s concurrently causing both the creaturely agent’s effect and its act. Thus, Suarez insists that “whatever is real must come immediately from God’s efficient causality; therefore, the causal activities and effects of all the [genera of] causes must come immediately from God’s efficient causality.”40 Moreover, given DUC, the whole of a creaturely effect and act must be caused by God, since no aspect of its being could exist without God’s immediately causing it. Yet, concurrentists typically hold that the entirety of a creature’s act and effect is brought about by its creaturely agent, as well. Thus, while any creaturely act involves the dual operation of God and the creature in bringing about the creaturely act and effect, this co-operation is not that of two partial contributors, each of which contributes only a portion of the effect jointly produced; it is rather the co-operation of two causes each of which produces the whole of the effect. As Aquinas instructs, “the same effect is not attributed to a natural cause and to divine power in such a way that it is partly done by God, and partly by the natural agent; rather it is wholly done by both.”41 And Suarez agrees that “it is not the case that part of the effect comes from the one cause and part from the other; rather, as St. Thomas notes … the whole effect comes from each.”42 Scholastic philosophers in the late Middle Ages and Renaissance worked out detailed accounts of how God concurs with creaturely operations. I will not pause to examine those details here.43 For present purposes, it is enough that we appreciate what DUC implies for God’s relationship to creaturely action. If DUC is true, then God causes every act and every entitative effect of any creaturely agent. It is these implications of the traditional doctrine that many contemporary theistic philosophers find problematic, especially for our understanding of creaturely freedom.44

1.2 Libertarian Freedom and the Apparent Conflict with DUC There is general agreement among philosophers on all sides of contemporary freewill debates that certain conditions must be met in order for an act or choice to be free.45 A free act must be voluntary, in the sense of something willed or desired by its agent. It must be intended by its agent—that is, done on purpose. As intended, it must be connected to the agent’s reasons for the sake of which it is done. And it must be under its agent’s control rather than something that merely happens to the agent. Closely connected with this last condition, a free act must be up to its agent and something for which the agent is genuinely responsible. One of the most significant disagreements within contemporary freewill debates is that between compatibilists and incompatibilists over whether these standard conditions for free action could be satisfied given a comprehensive determinism of all events (or at least of all events involving human beings).46 Compatibilists say “yes”; incompatibilists say “no.” In order to see what’s at issue between these rival camps, it is important to get clear on how “determinism” is understood within contemporary freewill debates. Robert Kane’s definition is representative:

6

Free Will and God’s Universal Causality: The Dual Sources Account An event … is determined when there are conditions obtaining earlier (such as the decrees of fate or the foreordaining acts of God or antecedent causes plus laws of nature) whose occurrence is a sufficient condition for the occurrence of the event. In other words, it must be the case that, if these earlier determining conditions obtain, then the determined event will occur.47

Determinism, thus, requires that there be a certain sort of relationship between any determined event, or determinatum, and its determinans, or thing determining it; namely, the determinans must be prior to the determinatum and must be a sufficient condition for the determinatum. Were this relationship not to obtain, then it would be improper to say that the one thing “determined” the other, at least in the sense of “determined” that figures in the dispute between compatibilists and incompatibilists. Two further points are needed to clarify the meaning of “determinism.” First, since the word “sufficient” has been used in multiple ways,48 it is important to emphasize that the sense of “sufficient” employed in the standard definition of “determinism” is logically sufficient.49 a is logically sufficient for b just in case it is not possible for a to exist (or occur, or obtain) without b’s existing (or occurring, or obtaining). Second, in defining determinism, writers like Kane frequently characterize the conditions that are sufficient for the determined event as being “earlier” than the event. I prefer “prior” to “earlier” as an adjective to characterize the conditions that are logically sufficient for a determined event. “Prior” can mean “temporally prior,” but “prior” is also frequently used to characterize one thing a’s relation to another b, when b asymmetrically depends on a, whether or not a temporally precedes b. Using this broad sense of “prior” in our definition of determinism allows us to say, for example, that an event is determined if it has causes that are logically sufficient for it, even if those causes do not precede the determined event in time. Indeed, what really matters for determinism is arguably not priority of time but priority in the order of dependence or explanation. With the clarification of “determinism” behind us, we can now see more precisely that what compatibilists and incompatibilists disagree about is whether free human action is compatible with all events having one or more factors that are both prior to and logically sufficient for them. By definition, compatibilism and incompatibilism are indifferent with respect to whether human beings actually perform free acts. Almost all contemporary compatibilists, in fact, think they do.50 Incompatibilists can easily be found on both sides of the question.51 Those incompatibilists who believe that we perform free acts are known as “libertarians.” A libertarian account of free will, thus, affirms human freedom and sees freedom as requiring that there be indeterminism at some relevant point in the process that produces a free act.52 Theists have typically affirmed human freedom, and in the debate between compatibilism and incompatibilism, a great many contemporary theists have sided with incompatibilism, thereby endorsing libertarian accounts of free will. Such theists have agreed with incompatibilists that if comprehensive determinism is true, and every event is determined, then no human act could satisfy all the conditions needed to be genuinely free. The conflict between freedom and determinism appears especially threatening when it comes to the requirement that a free act be “up to” its agent and something for which the agent is genuinely responsible.

Introduction

7

For instance, one might suppose that to be genuinely responsible for an act, an agent must have voluntarily performed the act, having had the ability voluntarily not to perform it, all prior conditions remaining the same. After all, how can I be responsible for an act if, given the antecedent conditions, it was not possible for me to do differently than I did? Yet if comprehensive determinism is true, then no human agent (at least not in the actual world) ever performs an act such that, given the prior conditions, the agent could have done differently. For, if comprehensive determinism is true, then every human act has prior conditions that are logically sufficient for it. But that means, for any act a human being performs, it was not possible for the agent not to perform that act, given those antecedent conditions. If an agent is, indeed, responsible for an act only if the agent could have done otherwise all antecedent conditions remaining the same, then no determined act is one for which an agent is responsible, and thus no determined act satisfies the conditions for a free act. Yet, perhaps this requirement for responsibility is too strong. Perhaps, an agent could be responsible for a determined act in virtue of being responsible for the antecedent conditions that determine it. Suppose, for example, that a person has developed such a habit of truth-telling that in a particular situation, given her character, it was simply impossible for her to tell a lie to avoid a minor embarrassment. Even if this person could not have done otherwise than remain truthful given the antecedent factors (especially the state of her character), if she is responsible for the state of her character, she is arguably also responsible for remaining truthful, despite the fact that she couldn’t have done otherwise.53 Yet, even granting that such cases are possible, and that in them the agent has responsibility for her determined act, it by no means follows that responsibility for her act is compatible with comprehensive determinism. An agent’s responsibility for her act in such cases derives from her responsibility for the antecedent conditions that determine her act. And she is responsible for these antecedent conditions in virtue of being responsible for some prior act (or refraining) from which these conditions result.54 Yet, not every act for which a human agent is responsible could be such that she is responsible for it in this derivative way. For, as we have just seen, every act for which an agent is derivatively responsible presupposes a prior act for which she is responsible. Thus, in order for all of a human being’s responsible acts to be derivatively responsible, she would have to perform an infinite number of prior acts for every one of her responsible acts. But no human being can do this, since every human being performs some first act, which takes place at some point after the person comes into existence. Consequently, for any act for which a human agent is derivatively responsible, there must be some act—even if the agent’s very first act—for which the agent is responsible in a basic or nonderivative way.55 Is such a basically responsible act possible given comprehensive determinism? Given comprehensive determinism, every human act is determined by antecedent conditions, but a basically responsible act is one for which the agent is responsible, even though she is not responsible for its antecedent conditions. But it seems doubtful that an agent could bear responsibility for an act determined by conditions for which she is not responsible. For, since the agent lacks responsibility for the determining conditions, these conditions are neither up to the agent nor within the agent’s control.

8

Free Will and God’s Universal Causality: The Dual Sources Account

And it is, presumably, also outside the agent’s control that these conditions are logically sufficient for the agent’s act. But, if neither the conditions nor the fact that the conditions are logically sufficient for the agent’s act is up to the agent, under the agent’s control, or something for which the agent is responsible,56 neither, it seems, is the act that follows as a necessary consequence of these conditions.57 In short, it looks doubtful that an agent can be responsible for an act that is determined by conditions not up to the agent or outside the agent’s control. Thus, no such act could be one from which an agent’s responsibility for other acts derives. And, so, it looks as if comprehensive determinism rules out basic responsibility and, therefore, also, derivative responsibility, since, as we have seen, derivatively responsible acts require at least one basically responsible act. If, then, an agent is responsible for performing an act that is determined, this responsibility must ultimately derive from her performing a basically responsible act, which is not determined. Thus, if all of our acts are determined, it appears that we can be responsible for none of them and hence that none of them is free. I do not pretend to have offered a thorough defense of the argument for incompatibilism, which task would require a lengthy consideration of compatibilist responses. Nevertheless, like a lot of contemporary theists, I find such arguments compelling, and am therefore attracted to libertarian accounts of human freedom. Unfortunately, it is far from obvious that libertarian human freedom is consistent with traditional theism given its commitment to God’s universal causality. For, as we saw in the previous section, DUC implies that every creaturely act is caused by God. Yet, according to prevailing opinion, if creaturely acts are caused by God, then they cannot be free in the libertarian sense. Representatives of this prevailing view are easy to find and worth quoting at length. In the course of presenting his free will defense, Alvin Plantinga, for instance, offers a decidedly incompatibilist or libertarian conception of human freedom: If a person S is free with respect to a given action, then he is free to perform that action and free to refrain; no causal laws and antecedent conditions determine either that he will perform the action, or that he will not.58

This conception, says Plantinga, rules out God’s causing a free creaturely act: If I am free with respect to an action A, then God does not bring it about or cause it to be the case either that I take or that I refrain from this action; … For if he brings it about or causes it to be the case that I take A, then I am not free to refrain from A, in which case I am not free with respect to A.59

Roderick Chisholm, similarly, recommends a libertarian account of responsibility: If the man was responsible for what he did, then, I would urge, what was to happen at the time of the shooting was something that was entirely up to the man himself. There was a moment at which it was true, both that he could have fired the shot and also that he could have refrained from firing it. And if this is so, then, even

Introduction

9

though he did fire it, he could have done something else instead. … But now if the act which he did perform was an act that was also in his power not to perform, then it could not have been caused or determined by any event that was not itself within his power either to bring about or not to bring about. For example, if what we say he did was really something that was brought about by a second man, one who forced his hand upon a trigger, say, or who, by means of hypnosis, compelled him to perform the act, then since the act was caused by the second man it was nothing that was within the power of the first man to prevent.60

Unfortunately for the libertarian proponent of DUC, the foregoing account of responsibility, says Chisholm, conflicts with a familiar view about the nature of God—with the view that St. Thomas Aquinas expresses by saying that ‘every movement both of the will and of nature proceeds from God as the Prime Mover’. If the act of the sinner did proceed from God as the Prime Mover, then God was in the position of the second agent we just discussed—the man who forced the trigger finger, or the hypnotist—and the sinner, so-called, was not responsible for what he did.61

Eleonore Stump and Norman Kretzmann express yet a similar judgment. In critiquing Brian Shanley’s interpretation of Aquinas, Stump and Kretzmann remark: For Aquinas, human beings have liberum arbitrium, and liberum arbitrium is the ability to do otherwise than one does: “we are said to have liberum arbitrium because we can take up one thing, having rejected another, which is what it is to choose” (ST 1.83.3). On Shanley’s interpretation of Aquinas as holding that God’s knowledge is causal, however, God causes everything (or everything temporal) that he knows. Since God knows human acts, on this view God’s knowledge causes those human acts. But if God causes human acts, then in what sense is it possible for any human being to act otherwise than she does? Clearly, it isn’t possible that God cause a person to do some act A and yet she does not-A. Is “ability to do otherwise” here supposed to have just the compatibilist sense of “ability to do otherwise if one chose to do otherwise”? On compatibilism, there is no genuinely open future for human beings because God determines the future in every respect. Surely, that is not the view Shanley means to attribute to Aquinas.62

Questions about the proper interpretation of Aquinas to one side,63 the passage indicates that, for Stump and Kretzmann, God’s causing human actions would require giving up libertarian human freedom and embracing compatibilism in its place. Jeffrey Brower agrees: “It is a contradiction to say that God causes an agent to freely choose (in the incompatibilist or libertarian sense) to perform some action.”64 This prevailing view—that God’s causing our acts precludes their being free in the libertarian sense—seems initially quite plausible.65 Drawing off the previous discussion, we can distinguish strict and broad accounts of what it is for an act to be free in the libertarian sense. According to the strict account, on which every free act

10

Free Will and God’s Universal Causality: The Dual Sources Account

must be one for which the agent could have done otherwise all antecedent conditions remaining the same: An act is free in the libertarian sense if and only if its agent performs the act voluntarily and intentionally, and the act is not determined (i.e., there is no factor both prior to and logically sufficient for the act).66

According to the broad account, which is happy to call “free” determined acts for which we are derivatively responsible: An act is free in the libertarian sense if and only if its agent performs the act voluntarily and intentionally, and either the act is not determined (i.e., there is no factor both prior to and logically sufficient for the act), or the act is determined and the agent’s responsibility for the act derives from the agent’s voluntary and intentional performance of some prior act that was not determined.

Given these accounts, God’s causing my acts precludes their being free in the libertarian sense if and only if God’s causing them either renders them determined or rules out their being performed voluntarily and intentionally. Now, my act being something I do voluntarily and intentionally would not appear in the least threatened by God’s causing my act. I know of no good reason to think that God couldn’t cause my doing something I do willingly (desirously) and on purpose. But, at least at first glance, God’s causing my act would seem to “determine” my act, introducing some condition—perhaps God’s act of causing my act or God’s willing or choosing to cause it—that is both prior to and logically sufficient for my act. If so, then God’s causing my acts rules out my ever having the ability to do otherwise all antecedent conditions remaining the same. Moreover, God’s causing my acts seems to make them ultimately up to God and not up to me. Indeed, it seems to preclude my acts being up to me, given the seemingly plausible assumptions both that God’s causing my act is not up to me and that it is not up to me that my act must occur if God causes it. In short, God’s causing all human acts would seem to introduce a comprehensive theological determinism, ruling out libertarian freedom and objectionable to libertarians on grounds that such determinism precludes human freedom and responsibility for precisely the same reasons given in the general argument for incompatibilism sketched above.

1.3 Dual Sources: A Neo-scholastic Approach to Resolving the Conflict If the prevailing view is correct, then those attracted to both traditional theism and libertarian freedom face an unwelcome choice: either abandon DUC, a central tenet of traditional theism, or give up on libertarian creaturely freedom. The present volume aims to show that, happily, the libertarian proponent of DUC can avoid this unwelcome choice. Contrary to the prevailing view, an act caused by God can still be free in the libertarian sense.67

Introduction

11

The approach defended in this book is not presented as that of any particular historical figure. Nevertheless, the approach can be characterized as “neoscholastic” (or perhaps broadly Thomistic) since many of the positions advanced here are inspired by, consonant with, and develop resources drawn from the scholastic tradition, especially Aquinas.68 Despite this orientation, the book is intended for contemporary thinkers of whatever stripe, but especially for analytic philosophers of religion and theologians. The account is recommended on its own merits and will be successful to the extent that it solves the problems it sets out to solve. The main aims of the book will naturally be of more interest to those who take DUC seriously. One reason to take it seriously is its central place within theistic tradition. In Chapter 2, I offer further grounds for taking it seriously by considering five arguments for DUC, an argument from scripture, and four philosophical arguments, including arguments from Anselm, Aquinas, and Suarez. Although these arguments will vary in their appeal depending on the presuppositions of the reader, I believe that for most theists, one or more of the arguments will be at least this compelling: It will give the theist reason to accept DUC provided that DUC does not preclude creaturely freedom or introduce other serious deficits. This modest conclusion is significant because it indicates the desirableness of accepting DUC, and thus motivates my project, which attempts to remove the chief obstacles to doing so. One obstacle to accepting DUC, which is more general than the concern that it rules out libertarian freedom, is the concern that it rules out productive creaturely agency of any sort. Recall that, according to DUC, the whole of whatever exists distinct from God is immediately caused by God. It follows that the whole of any creaturely effect is immediately caused by him. As we’ve seen, scholastic philosophers such as Aquinas and Suarez held that the whole of a creaturely effect is caused by both God and the creature. Yet, a number of contemporary philosophers have argued that God’s causing the whole of a creaturely effect makes the putative creaturely cause superfluous or otiose. If such philosophers are correct, the theist must either abandon DUC or embrace occasionalism, the radical and, to many, objectionable denial that creatures are efficacious. In Chapter 3, after elaborating on the relationship between God and creaturely causes, I consider arguments for the claim that DUC rules out productive creaturely agency and show that none of these arguments succeeds. Not only can one embrace DUC without occasionalism; more specifically, one can embrace DUC alongside an agent-causal understanding of free creaturely agency. Chapters 4 and 5 are pivotal. Key to reconciling DUC with libertarian freedom is developing an account of divine agency on which God’s causing a creaturely act does not render the act “determined.” Aquinas points in the direction of this strategy when, at Summa theologiae 1–2.10.4, he denies that God’s moving the will necessitates it: As Dionysius says (Div. Nom. iv) it belongs to Divine providence, not to destroy but to preserve the nature of things. Wherefore it moves all things in accordance with their conditions; so that from necessary causes through the Divine motion, effects follow of necessity; but from contingent causes, effects follow contingently. Since, therefore, the will is an active principle, not determinate to one thing, but having an indifferent relation to many things, God so moves it, that He does not

12

Free Will and God’s Universal Causality: The Dual Sources Account determine it of necessity to one thing, but its movement remains contingent and not necessary, except in those things to which it is moved naturally.69

As far as I know, Aquinas never explains in detail how it can be that God’s moving the will or causing a creaturely act doesn’t determine or necessitate it. But, focusing on the sense of “determinism” relevant to contemporary freewill debates, it is clear that what’s needed is an account of divine agency on which God’s causing a creaturely act introduces no factor that is both prior to and logically sufficient for the act. In Chapter 4, I present such an account of divine agency—what I call the extrinsic model of divine agency—and show that, given this model, it is possible for an act caused by God to be free in the libertarian sense. I also explain how the solution I offer develops a Thomistic tradition that identifies the divine transcendence as the key to reconciling human freedom and divine causality. As we will see, not only can an act caused by God satisfy both the strict and broad accounts of what it is for an act to be free in the libertarian sense. The extrinsic model also enables an act caused by God to satisfy those requirements for free agency that libertarians (and incompatibilists, generally) believe to be ruled out by comprehensive determinism. Thus, it will be shown that an act caused by God can be one for which its creaturely agent had the ability to do otherwise all antecedent conditions remaining the same. And it will be shown, also, that an act caused by God can be genuinely “up to” its creaturely agent, within its agent’s “control,” and something for which the agent bears “ultimate responsibility,” as these concepts have been understood by libertarians. Because an act caused by God can still be “up to” its creaturely agent, and something for which its agent bears ultimate responsibility, the extrinsic model enables us to say that a free creaturely act, while caused by God, nevertheless, has its source in its creaturely agent, as well. On the account that will emerge, a free creaturely act has dual sources— God and its creaturely agent—and is genuinely “up to” both. Such, I think, is precisely what the libertarian proponent of DUC should want to say. It is for this reason that I refer to my overall approach as the “Dual Sources account” or “Dual Sources.” While Chapter 4 presents the extrinsic model of divine agency and explains how it renders DUC compatible with libertarian freedom, Chapter 5 offers an extensive defense of the extrinsic model. I show that the model is not ad hoc, constructed merely for the purpose of reconciling libertarian freedom with DUC. On the contrary, the extrinsic model is an integral part of scholastic theology, since it is the model best able to accommodate the scholastic understanding of God’s relations to creatures and scholastic commitment to both divine simplicity and divine freedom. Moreover, the extrinsic model is arguably implied by DUC itself, together with a handful of assumptions, all of which are plausible, and many of which are widely shared by theists, whether scholastic or not. The extrinsic model, therefore, is an account of divine agency that a proponent of DUC arguably must embrace anyway, even apart from its usefulness in reconciling DUC with libertarian freedom. As we saw in the previous section, Chisholm understandably judged that God’s causing an act of sin would relieve the sinner of responsibility. If the argument of Chapter 4 succeeds, then Chisholm’s worry disappears, since an act caused by God can still be one for which its creaturely agent is responsible. A problem remains, however,

Introduction

13

since most theists wish not only to secure the sinner’s responsibility for sin but also to distance God from responsibility. Yet, if God causes sinful acts, it looks as if God causes sin every bit as much as the sinner. Chapter 6 takes up this difficult problem. Following the strategy of authors such as Anselm and Aquinas, I argue that a sin of commission is not just an act but an act with a defect or privation in virtue of which it is sinful. To cause the sin requires causing both the act and the defect. While the sinner causes both, God causes only the act, and therefore does not cause the sin. A number of formidable objections have been raised against this strategy, but I maintain that all of them can be answered. Even if God does not cause sin, it might still be wondered how God could be justified in permitting it. Here we come to the problem of moral evil understood as a challenge to the claim that the amount and types of moral evil we find in the world are consistent with the existence of an all-powerful, all-good God. In Chapter 7, I argue that the most popular element in contemporary responses to the problem of moral evil—the Free Will Defense (FWD)—fails. According to FWD, God was confronted with a choice between mutually exclusive alternatives: He could either grant some of his creatures the libertarian freedom to perform or refrain from morally bad acts, or he could ensure that no morally bad acts ever occur. He could not do both. FWD falters, I argue, because Dual Sources shows how God’s giving us libertarian freedom to perform or refrain from morally bad acts is perfectly consistent with God’s ensuring that we always choose the good. This unexpected result may seem an unhappy one. But I go on to argue, in Chapter 7, that the loss of FWD is not as devastating as it may at first appear. For the theist can respond to the problem of moral evil without FWD. Indeed, I argue that, in accepting Dual Sources and rejecting FWD, the theist can offer as strong (or nearly as strong) a response to the problem of moral evil as he could if Molinism or Open Theism—the two most popular alternative approaches for combining theism and libertarian freedom—were true. William Hasker once wrote, in a way that still represents the dominant view, that “if you wish to maintain a strong doctrine of providence … and yet to uphold libertarian free will, then Molinism is the only game in town.”70 Chapter 8 shows that there is another game, explaining how Dual Sources enables robust accounts of God’s foreknowledge, providence, grace, and predestination, all consistent with libertarian creaturely freedom. Although the book does not spend much time discussing and critiquing other approaches, by the end it will be clear that Dual Sources offers a comprehensive alternative to the currently most popular approaches for combining theism with libertarian freedom, in particular to Molinism and Open Theism, with which I will assume the reader’s familiarity. In common with Molinism, but contrary to Open Theism, Dual Sources understands God’s providence to extend in detail to all that happens, including free creaturely actions. Again, like Molinism, but contrary to Open Theism, Dual Sources affirms God’s exhaustive knowledge of the whole of history, including of what creatures freely do in the future. Unlike Molinism, however, Dual Sources does not depend on the doctrine of middle knowledge, that is, on God’s knowledge, prior to any creative act on his part, of truths about what any possible free creature would freely do in any possible circumstance in which it might find itself. The Dual Sources approach may, therefore, be especially attractive to those who doubt that

14

Free Will and God’s Universal Causality: The Dual Sources Account

there are such truths for God to know, but who nevertheless wish to affirm libertarian creaturely freedom alongside a traditional understanding of God’s providence and knowledge. Dual Sources will likely also be attractive to those who would like to pair libertarian creaturely freedom with a strong doctrine of God’s grace, one on which not just the possibility of our acting meritoriously but our meritorious actions themselves, and our salvation (if we be so blessed), are God’s gracious gifts.71 The issues taken up in this book are among the most disputed in the history of theology and philosophy of religion. All the comprehensive accounts of God’s relationship to creaturely freedom have benefits and costs that have variously attracted or deterred potential adherents. While the Dual Sources account will be no different, it does constitute a largely new or overlooked approach among contemporary philosophers of religion. This book aims to develop the approach and to make the case that it deserves serious consideration alongside the alternatives.

2

God: Universal Cause and Cause of Human Actions

The doctrine of divine universal causality (DUC) holds that necessarily, for any entity distinct from God, God directly causes that entity to exist at any time it exists. It follows from DUC that creaturely acts are directly caused by God. This book does not attempt to establish DUC. Nevertheless, the book’s central thesis—that our acts can be caused by God and still be free in the libertarian sense—is of considerably more interest to the extent that there is reason to take DUC and the claim that our acts are caused by God seriously. Of course, one reason to take these claims seriously is their central place within traditional theism, documented in the last chapter (Section 1.1). The current chapter moves beyond a presentation of what theists have traditionally believed to gather reasons that have been given in support of DUC and of its corollary that creaturely acts are caused by God. Beginning with support found in scripture, the chapter goes on to survey a variety of philosophical arguments for these claims, including arguments by three figures important to the scholastic tradition: Anselm (1033–1109), Aquinas (1225–1274), and Suarez (1548–1617). Because the arguments considered start from such different presuppositions, the appeal of this or that argument will vary with the presuppositions of the reader. What can hardly be denied is that DUC, with its implications for God’s relationship to creaturely action, has been favored for a variety of reasons. For most theists, one or another of these reasons should be at least this compelling: It should give the theist reason to embrace DUC provided the doctrine can be reconciled with other things the theist wants to hold.

2.1 Scripture The Old Testament declares God creator of heaven and earth (Gen. 1–2) and maker of all things (Isa. 44:24). The New Testament echoes this conviction, seeing all things as having been made through the Word, who was with God from the beginning and who is God (Jn 1:1–3).1 Although such passages are by themselves consistent with God’s causal role being limited to getting the world started, many other passages make clear that God is the source of things absolutely speaking and not simply at the beginning. Thus, Hebrews proclaims God to be “for whom and through whom all things exist” (Heb. 2:10). 1 Corinthians states that “all things come from God” (1 Cor. 11:12),

16

Free Will and God’s Universal Causality: The Dual Sources Account

teaching that “there is one God, the Father, from whom are all things … and one Lord, Jesus Christ, through whom are all things” (1 Cor. 8:6). “O the depth of the riches and wisdom and knowledge of God!” declares Romans, “for from him and through him and to him are all things” (Rom. 11:33, 36).2 To say that all things exist “by” and “through” God suggests a doctrine of divine conservation—that things depend on God not only for their beginning to exist but for their existing at any time. God’s conserving of creatures is also suggested in St. Paul’s famous address at the Areopagus: Athenians, I see how extremely religious you are in every way. For as I went through the city and looked carefully at the objects of your worship, I found among them an altar with the inscription, “To an unknown god.” What therefore you worship as unknown, this I proclaim to you. The God who made the world and everything in it … He himself gives to all mortals life and breath and all things. From one ancestor he made all nations to inhabit the whole earth, and he allotted the times of their existence and the boundaries of the places where they would live, so that they would search for God and perhaps grope for him and find him—though indeed he is not far from each one of us. For “In him we live and move and have our being.” (Acts 17:22–28)

One who gives us life and breath and in whom we live and move and have our being is not one for whom we depend merely for our beginning to exist. Nor does the language of these passages lend itself to the thought that some things depend on God only, for instance, by receiving their existence from some third thing, which receives its existence from God. On the contrary, the prepositions “by,” “through,” and “in” suggest that God himself is more directly the source of all things’ existence and thus that all things depend on God immediately. If we take “all things” in these passages to refer quite literally to all entities distinct from (i.e., not identical to) God, then we arguably find the bulk of DUC therein affirmed. But, perhaps, “all things” should be taken more narrowly to refer only to creaturely substances? On such a reading, the foregoing passages would not imply, as DUC does, that God brings about creaturely operations. While this narrow reading cannot be decisively ruled out when the foregoing passages are taken in isolation, scripture as a whole depicts God’s contribution as far more extensive. Not only does God bring forth and conserve creaturely substances together with the powers by which they perform; God is actively at work in his creatures’ performances. As the late, prominent Old Testament scholar Walther Eichrodt observes:3 It was possible [for Old Testament authors] … spontaneously and axiomatically to portray natural events … as a direct act of God, who controls both Nature and history by the omnipotence with which he fills all things. Thus the bestowal of rain and fertility is the direct gift of Yahweh; in the rain he blesses the field,4 and causes the plants to grow, and gives the animals their food5; he nourishes his people with fruits of the ground,6 and punishes her by withholding them.7 Just as he desires to be petitioned afresh every time for the blessing of children, which he grants or

Universal Cause and Cause of Human Acts

17

refuses,8 so he himself forms the individual human being in his mother’s womb, clothes him with skin and flesh, gives him the breath of life, and summons each new generation into existence.9 The sick man recognizes his sickness as God’s visitation,10 the shipwrecked sailor sees him stirring the sea by means of the storm wind.11 The earthquake comes from the blow of his fist,12 the smoking volcanoes have felt his touch.13 After the night he brings in the morning,14 and holds sway over the course of the stars.15 The regular return both of the seasons of the year with their gifts of harvest and of the comforting light of day after the dark of the night spring from the paternal care of the Creator, who even after the judgment of the Deluge wills to preserve his creature.16 In everything his wonders may be discerned17; the pious soul, overleaping all intermediate causes, sees God forming the universe at every moment.

Such examples testify to the conviction that God is at work in the operations of nature; yet this conviction is no less evident when it comes to the volitional acts of human beings. Consider that in scripture the human “heart” represents the will or center of volition, the place of moral choice.18 Yet, we learn from Proverbs that “the king’s heart is a stream of water in the hand of the Lord; he turns it wherever he will” (Prov. 21:1). While the Psalmist declares: “The Lord looks from heaven, he sees all humankind; from where he sits enthroned he watches all the inhabitants of the earth, he who fashions the hearts of them all” (Ps. 33:13–15). According to Acts 17:28, we not only “live” and “have our being” in God, but we also “move” in him. Proverbs teaches that “all our steps are ordered by the Lord” (Prov. 20:24), while Isaiah says, “O Lord, … all that we have done, you have done for us” (Isa. 26:12).19 In Ezekiel, God proclaims to his people: “I will put my spirit within you, and make you follow my statutes and be careful to observe my ordinances” (Ezek. 36:27). Philippians teaches, “It is God who is at work in you, enabling you both to will and to work for his good pleasure” (Phil. 2:13).20 And 1 Corinthians holds that “there are varieties of activities, but it is the same God who activates all of them in everyone” (1 Cor. 12:6). Note that it is not only good creaturely deeds but also those of the likes of Herod and Pilate, which have been, according to Acts 4:23–28, planned and ordained by God. Consider, also, Isaiah 63:17, where God is credited with being causally behind his creatures’ behavior, even when creatures stray from his ways: “Why, O Lord, do you make us stray from your ways and harden our heart, so that we do not fear you?” Given that God operates in the operations of both nature and will, we should not be surprised that scripture depicts God as behind all events and sovereign over the course of history. Judith gives especially vivid expression to this picture: O God, my God … you have done these things and those that went before and those that followed. You have designed the things that are now, and those that are to come. What you had in mind has happened; the things you decided on presented themselves and said, ‘Here we are!’ (Jdt. 9:4–6)

God’s plan, it is suggested, cannot be thwarted:

18

Free Will and God’s Universal Causality: The Dual Sources Account This is the plan that is planned concerning the whole earth; and this is the hand that is stretched out over all nations. For the Lord of hosts has planned, and who will annul it? His hand is stretched out, and who will turn it back? (Isa. 14:26–27)

God’s sovereignty extends even to chance events: The lot is cast into the lap, but the decision is the Lord’s alone. (Prov. 16:33)

So also are the life stories of human beings given shape by their heavenly author: Your eyes beheld my unformed substance; in your book were written all the days that were formed for me, when none of them as yet existed. (Ps. 139: 16)

Thus, humans are compared to instruments in God’s hands (Isa. 10.15) and to clay in the hands of the heavenly potter (Isa. 29:16, 45:9, 64:7; Jer. 18:6; Rom. 9). One may reasonably dispute whether a given passage from those referenced in the preceding paragraphs should be taken to teach that God causes all human actions. For instance, perhaps 1 Cor. 12:6, quoted two paragraphs above, applies only to actions that manifest gifts of the Spirit; and some may even wish to argue that God’s “activating” such actions involves only his giving the spiritual gifts and not also his causing the actions performed under the influence of these gifts. Nevertheless, not all the passages that speak of God’s causal relation to human action plausibly admit of such a restricted reading. And when these passages are taken together, and when they are read alongside those passages that represent God as the cause of “all things,” and as causally operative in the operations of nature, and as the sovereign designer of the entire course of history extending even to human and chance events, they not only suggest a theology on which God is causally at work in all creaturely operations (both natural and free); they also recommend DUC as a doctrine the acceptance of which affords a most economical means of accounting for the totality of this biblical evidence regarding God’s causal relationship to everything else. There are, of course, scriptural passages that appear to present a different picture. For example, it may seem that God’s plan is indeed thwarted in cases, such as Adam and Eve (Gen. 3), and Jonah (Jon. 1:1–3), where his creatures disobey his commands. Furthermore, God is sometimes portrayed as ignorant of his creatures’ behavior (Gen. 18:20–21) or as surprised by it (Jer. 3:7, 32:35), neither of which we would expect of a God who causes that very behavior.21 The seeming tensions between these passages and passages of the sort referenced above may suggest that it is a mistake to look to scripture for a coherent depiction of God’s relationship to creaturely action. The authors of scripture, after all, were many and the contexts and purposes of their writings various. Why think there is a consistent theology to be found? Cautionary notes of the sort just expressed make good sense where scripture is viewed as no more than a collection of human writings. Such resignation is deeply unsatisfactory, however, for those who believe that behind the many human authors is the single Divine Author who uses scripture to reveal himself. Such persons will naturally look to scripture for coherent teaching. They will, therefore, need some

Universal Cause and Cause of Human Acts

19

interpretive strategy to avoid reading scripture as affirming or implying inconsistent propositions. Where two passages seem to conflict, they will need a plausible interpretation of what both affirm that renders the passages compatible. Could it be that, properly interpreted, the many passages that appear to affirm God as causally behind the operations of creatures, in fact, make no such affirmation? The suggestion seems less likely than the alternative—that, properly understood, the passages in supposed tension with this first set do not, in fact, deny that God causally operates within his creatures’ actions. Take, first, cases of disobedience. From the fact that a creature disobeys God’s command, it doesn’t follow as a logical implication that God does not cause the creature’s act of disobedience or that the disobedience is not included under God’s plan. What is more, on more than one occasion, scripture actually depicts God as causally operative in such disobedience. As we have seen, Isaiah asks, “Why, O Lord, do you make us stray from your ways and harden our heart, so that we do not fear you” (Isa. 63.17)? And there is the (in)famous case of Pharaoh: And the Lord said to Moses, “When you go back to Egypt, see that you perform before Pharaoh all the wonders which I have put in your power; but I will harden his heart, so that he will not let the people go. Then you shall say to Pharaoh, ‘Thus says the Lord: Israel is my firstborn son. I said to you, ‘Let My son go that he may worship me.’ But you refused to let him go; now I will kill your firstborn son.” (Exod. 4.21–23)

Such passages raise all sorts of questions. The point here is simply that from scripture’s presenting certain acts as acts of disobeying God, or acts of which God disapproves, we cannot conclude that scripture denies that God is causally behind such actions.22 Nor should passages that portray God as ignorant of, or surprised by, his creatures’ behavior be read as implying that God does not cause that behavior. For on the assumption that scripture offers coherent teaching, there is good reason to think we should not take such portrayals of divine ignorance literally. Elsewhere scripture indicates that God sees everything under the heavens (Job 28:24), observes all our deeds (Ps. 33:13–15), sees our ways (Job 31:4), and beholds all our steps (Job 34:21). Not only what is presently happening but also the future God knows. So the Psalmist declares, “Even before a word is on my tongue, O Lord, you know it completely … in your book were written all the days that were formed for me, when none of them as yet existed” (Ps. 139:4, 16); while Jesus foreknows the betrayal of Judas (Mk 14.18–20) and the denials of Peter (Mk 14.27–30).23 Scripture offers many passages in which God discloses, by himself or through a prophet, what is to come.24 Knowledge of the future is even presented in Isaiah as a criterion for distinguishing the true God from idols.25 Of course, some looking to scripture for coherent teaching insist that the portrayals of divine ignorance should be taken literally, and that what should be denied is that passages such as those in the previous paragraph really affirm or imply God’s complete knowledge of creaturely action, including future creaturely action. Such persons must also deny that scripture teaches that God

20

Free Will and God’s Universal Causality: The Dual Sources Account

causes creaturely operations (unless they want to maintain that God’s so causing is somehow consistent with his being ignorant of what his creatures are doing, as some passages in scripture present him to be). Space does not permit an examination of the interpretive strategies such persons use in an effort to dispel the impression that scripture has God causing all creaturely operations and knowing all—even future—creaturely actions. Here I simply declare my judgment that these interpretive strategies are less plausible than simply denying that the portrayals of divine ignorance be taken literally.26 There is nothing new or difficult in the idea that scripture characterizes God in ways that are not literally true for the sake of conveying some message that is true. God is not literally a rock, a fortress, or a shepherd, but characterizing him as such is a compelling way of communicating the truths that God is strong and that he protects and cares for his people. The logic used when interpreting biblical metaphors can easily be extended to certain biblical narratives. Consider these: Then the Lord said: “How great is the outcry against Sodom and Gomorrah and how very grave their sin! I must go down and see whether they have done altogether according to the outcry that has come to me; and if not, I will know.” (Gen. 18:20–21) This word came to Jeremiah from the Lord: Take a scroll and write on it all the words I have spoken to you against Israel and Judah and all the nations … It may be that when the house of Judah hears of all the disasters that I intend to do to them, all of them may turn from their evil ways, so that I may forgive their iniquity and their sin. (Jer. 36: 1–3)

Taken literally, the first passage implies that God lacks complete knowledge of the present, and, seemingly also, the past. The second passage suggests that God lacks complete knowledge of the future.27 Yet, the dominant scriptural testimony suggests that we not take these literally. Rather, just as calling God a fortress is a literally false yet compelling way of communicating the truth that God protects his people, so the foregoing passages are a literally false yet compelling way of communicating the truth that human decision and behavior are of critical importance. By portraying God in a narrative context as surprised by what his creatures do, or as wondering what they have done, are doing, or will do—as if he didn’t know—these passages place the dramatic focus on human action. They thereby convey the truth that much depends on what we do. Yet, since the portrayals of divine ignorance are not to be taken literally, they do not any more than the cases of disobedience undermine the conclusion that scripture has God causally behind human actions.28 Scripture, of course, abundantly testifies to the significance of human freedom and responsibility.29 Indeed, I suspect that many of those who deny that scripture teaches that God causes human actions, or that God knows what humans will freely do in the future, would not find these denials plausible if they did not believe these claims to be incompatible with the human freedom and responsibility scripture so clearly affirms. They reason, I suspect, as follows:

Universal Cause and Cause of Human Acts

21

1. Scripture teaches that humans enjoy significant freedom and responsibility. 2. If scripture teaches some proposition p, then scripture does not teach some proposition q that is inconsistent with p. 3. The proposition that God causes human actions, or knows what humans will do, is inconsistent with the proposition that humans enjoy significant freedom and responsibility. Therefore, 4. It is not the case that scripture teaches that God causes human action or knows what humans will do. If my suspicion is correct, then the judgment that (4) very often rests crucially on the extra-scriptural judgment that (3) and would not be reached without it. But this suggests that a purely scriptural case for (4) is weak or, at the very least, underdetermined. In fact, if we read scripture without the extra-scriptural judgment that (3), it is more natural to affirm the negation of (4) from which, with (1) and (2), we could infer the negation of (3).30 Citing many passages of the sort we have looked at in this section, Eichrodt maintains: From the very beginning Israel’s belief in God’s power in history and individual life went far beyond that large-scale fulfillment of his decrees which made use even of human refractoriness for its own ends, and implemented its royal dominion in the teeth of all external opposition. Even the innermost life of Man was subjected to the all-pervading divine energy. It is not simply that God allows a man to think thus and not otherwise; he is himself also at work within these acts of personal freedom. … One will never do justice to the profound grasp of the reality of God which is evinced in these statements by trying to explain them in terms of God’s permissive will. … In the cases mentioned what is involved is a real act of God, in whose hand men are as clay in the hand of the potter.31

At the same time, says Eichrodt, the Old Testament’s emphasis on God’s causal presence within human action never led to a flat determinism, depriving Man of the responsibility for his actions. At all times the capacity for self-determination is insistently retained. The whole ethical exhortation of the prophets is based on the conviction that decision is placed in the hands of men. But the Law, too, setting before Man the choice of life or death, rests on this presupposition. The fundamental postulate of moral freedom is found in equal force alongside the religious conviction of God’s effective action in all things; and no attempt was made to create a harmonizing adjustment between them. It is testimony to the compelling power of the Old Testament experience of God that it was able to affirm both realities at once, and to endure the tension between them, without discounting anything of their unconditional validity.32

New Testament scholar John M.G. Barclay finds similar affirmations in Paul’s epistles, remarking:

22

Free Will and God’s Universal Causality: The Dual Sources Account Some of Paul’s statements on the relation of divine to human agency strike us as paradoxical, if not downright incoherent. He urges the Philippians, for instance: “Work out your own salvation with fear and trembling, for God is at work in you, both to will and to work for his good pleasure” (Phil. 2.12–13). In statements such as this (cf. 1 Cor. 15.9–10) we are perplexed by the juxtaposition of two agencies: that of the Philippians, the recipient of Paul’s exhortation, who are clearly responsible for their “work”, and that of God, whose “work” is taken to be not independent of theirs, but in some sense the source of their action, even of their will to act.33

Dual Sources aims to show that this fundamental outlook of scripture is, in fact, coherent—that human beings can have morally significant, self-determining freedom, even while all creaturely acts are caused by God.

2.2 Perfect Being Theology: An Anselmian Approach The previous section considered arguments from scripture for DUC and for the corollary that human acts are caused by God. In the remainder of this chapter, I will consider various philosophical arguments for these theses, beginning with an argument based in the methodology known as “perfect being theology.” How should we characterize God? Perfect being theologians suggest that we think about what qualities would belong to the greatest or most perfect being possible, and whatever these are, God has them. The father of this approach is St. Anselm, who in his Proslogion famously maintained that the very idea of God is of a being “than which none greater can be thought.”34 From this starting point, Anselm reasons that God must be “whatever it is better to be than not to be,” for of anything less than this, something greater could be thought and so anything less would not be God.35 Anselm applies his method in the Proslogion to derive a number of the traditional divine attributes, that God is living, perceptive, truthful, good, wise, just, merciful, omnipotent, blessed, eternal, impassible, simple, and so forth.36 Perfect being methodology enjoys a number of contemporary defenders,37 and even apart from its declared devotees, it is probably the most popular approach for thinking about the divine attributes aside from approaches that make explicit appeal to divine revelation. Is there an argument for DUC using perfect being methodology? Anselm thought so. The heading for Proslogion Chapter 5 reads: “That God is whatever it is better to be than not to be and that, existing through Himself alone, He makes all other beings from nothing.”38 This brief chapter begins: What then are You, Lord God, You than whom nothing greater can be thought? But what are You save that supreme being, existing through Yourself alone, who made everything else from nothing? For whatever is not this is less than that which can be thought of; but this cannot be thought about You.39

It might be argued that by “all other beings” and “everything else” Anselm intends only all substances distinct from God and not also the properties and actions of substances.

Universal Cause and Cause of Human Acts

23

But we know from Section 1.1 that Anselm takes God to be the cause of these things too precisely because they are real and distinct from God. Thus, we may comfortably suppose that by “all other beings” Anselm intends literally all that exists apart from God, including creaturely actions. Contemporary perfect being theologian Thomas Morris reasons similarly to Anselm. In presenting a conception of the greatest possible being, he describes God as a thoroughly benevolent conscious agent with unlimited knowledge and power who is the necessarily existent, ontologically independent creative source of all else.40

As Morris puts it, “a being of unlimited power and knowledge who was the source of all other beings would seem to be superior to one who, for all his excellence, was just one among other independent beings.”41 William Lane Craig concurs: Being the cause of existence of other things is plausibly a great-making property, and, the maximal degree of this property is to be the cause of everything else that exists. God would be diminished in His greatness if He were the cause of only some of the other things that exist. … God’s greatness would be even further augmented if it were impossible that anything exist independent of His creative power. Thus, in any possible world God, as the greatest conceivable being, is the source of all things, if any, apart from Himself.42

Could a practitioner of perfect being theology resist the inference to God’s universal causality? A potential drawback of the method is that its results depend so critically on the practitioner’s intuitions about what properties are truly perfective or great-making. For example, while classical practitioners such as Anselm thought that a perfect being would be immutable, Richard Swinburne wonders why the “perfection of a perfect being might [not] consist … in his being in a certain process of change,” suggesting that “only neo-Platonic dogma would lead us to suppose otherwise.”43 Might a perfect being theologian similarly deny that “being cause of all that exists distinct from oneself ” is truly great-making? While I don’t see how the methodology rules out such a denial, I do think it is highly plausible that a being that causes all that exists apart from itself is, all things being equal, at least of greater metaphysical stature than a being that does not. The most promising way for a perfect being theologian to resist God’s universal causality, then, is not to deny that “being cause of all else” is a great-making property, but rather to argue that this property is incompatible with some other property that it is even greater to have.44 How might such an argument go? In my view, the most plausible version would maintain that DUC, with its implication that God causes creaturely action, is incompatible with God’s having the power to create beings with morally significant—let us assume libertarian—freedom. It would then argue that having the power to create beings with libertarian freedom is greater than being cause of everything apart from oneself and therefore that DUC is false.45 Similarly, one might argue that if DUC is true, then God causes sin. But God’s

24

Free Will and God’s Universal Causality: The Dual Sources Account

causing sin is incompatible with the goodness of a perfect being, a goodness that it is greater to have than being cause of all that is not oneself. And, so again, DUC should be rejected. Our dialectical situation has become rather interesting. Among the main arguments of the present book are that creaturely acts caused by God can still be free in the libertarian sense and that God can even cause acts of sin without causing sin itself or compromising his goodness. The purpose of this chapter is to motivate the project by looking at reasons to take DUC seriously. We have seen that there is a strong case, using perfect being methodology, for God’s universal causality. The most significant threat to this case turns on the bet that the main arguments of the book fail. If, however, Dual Sources succeeds in showing that libertarian freedom is compatible with DUC, and that God’s causing the act of sin neither makes God cause of sin nor compromises his goodness, then the perfect being theologian has good reason to affirm both DUC and its implications for God’s relationship to creaturely action.

2.3 Cosmological Arguments from Contingency Cosmological arguments from contingency have traditionally been among the most popular arguments for God’s existence. Such arguments reason to God as a necessary being needed to account for the existence of contingent beings.46 Of particular relevance to our question are what I will call “conjunctive arguments from contingency.” These arguments do not merely conclude (as do some arguments from contingency) that there would be nothing contingent were it not for the activity of a necessary being. Rather, conjunctive arguments infer the existence of God to account for the whole totality, conjunction, or aggregate of contingent entities. Though such arguments vary in their details, what they have in common is the conclusion that every contingent entity has God as its cause. Conjunctive arguments have been advanced, or at least appear to have been, by some of the most able recent defenders of cosmological arguments. Thus, in presenting their “new cosmological argument,” Gale and Pruss define our world’s “Big Conjunctive Contingent Fact” as the conjunction of all contingent propositions that are true.47 They then state their intent to establish a “creator of this world’s universe, where a world’s universe is what verifies or makes true all of the conjuncts in this world’s Big Conjunctive Contingent Fact.”48 Like Gale and Pruss, Koons defends an argument in which the whole aggregate of contingent facts has God as its cause.49 Similarly, O’Connor reasons to God as the cause of the whole contingent order, “the enormously complex range of connected events involving the existence and interactions of concrete objects and systems.”50 According to O’Connor: The existence of each natural particular and the events in which they participate admit, in principle, of a fully adequate explanation in terms ultimately involving their causal dependency on a necessary being, whose activity was guided but not determined by some goal(s) that the actual order of things were seen to satisfy.51

Universal Cause and Cause of Human Acts

25

My purpose is not to evaluate conjunctive arguments from contingency but rather to note their implications. Assuming they establish a single necessary being that brings about the entire totality or collection of contingent entities,52 they imply that God causes all human actions, since human actions are members of that collection. Though this latter implication is not explicitly mentioned by the authors cited above, it has been noted by others who have advanced conjunctive arguments. Thus, Herbert McCabe remarks: If what I have been saying [in my cosmological argument] is true then we must conclude … that since everything that exists owes its existence to God, since he is the source of anything being rather than nothing, he must also be the source of my free actions.53

Similarly, Germain Grisez notes that his cosmological argument raises “the problem of the compatibility between man’s action being free and its being within the scope of the creator’s uncaused causing.”54 Another reason to take seriously the claim that our acts are caused by God, then, is that it is an immediate implication of a type of argument for God’s existence that enjoys a number of prominent defenders. Anyone who endorses such an argument would appear committed to our acts being caused by God.

2.4 Conservation and Concurrence: A Suarezian Argument As we know from Section 1.1, Suarez holds that every creaturely operation requires God’s concurrence—that is, that God co-operate with the creature by causing both the act and effect of the creaturely agent. In Metaphysical Disputations 22.1, Suarez advances several lines of argument for this position.55 One argument is quite similar to that made when discussing perfect being theology. Assuming God’s concurrence with creaturely operations is possible, a God who so acted would exhibit far greater power and more perfect sovereignty than a God who did not: This manner of acting in and with all agents pertains to the breadth of divine power, and on God’s part presupposes a perfection untainted by imperfection; … There is in this way a perfect and essential ordering between the First Cause and the secondary cause, and there is nothing impossible here … Therefore, this general influence should not be denied to God.56

Suppose, then, that the arguments given in the remainder of this book succeed and that God’s causing creaturely operations is shown to be coherent and even consistent with libertarian freedom. In Suarez’s view, we would have every reason to affirm the concurrentist thesis that creatures act only insofar as their actions and effects are caused by God. Perhaps the most arresting of Suarez’s philosophical arguments for concurrence, however, is his contention that the doctrines of divine concurrence and divine

26

Free Will and God’s Universal Causality: The Dual Sources Account

conservation have the same underlying rationale, such that if one rejects the need for concurrence, one ought also reject the need for conservation. In Suarez’s words: If it is not the case that all things are effected immediately by God, then neither is it the case that they are conserved immediately.57

Dialectically, this argument is potentially powerful, since theists who reject divine concurrence nevertheless typically affirm divine conservation. Consider these statements from some prominent contemporary philosophical theists, all of whom (I believe) either are silent about or reject concurrence: van Inwagen: “I hold … that no created thing could possibly exist at a given moment unless it were at that moment held in existence by God; and no created thing could possibly have causal powers at a given moment unless it were at that moment supplied with those powers by God.”58 Plantinga: “We depend upon God’s creative and sustaining activity for our existence. … It is not possible, at least if traditional theism is correct, that we should exist and God not create and sustain us.”59 Morris: “All things distinct from God stand in a dependence relation to God, a relation that is both direct and absolute. It is never the case that some created object x depends upon God only in the sense of depending for its existence upon some other created objects y and z, which in turn directly depend on God. Every created object depends directly on God for its existence. … The dependence is thorough and continuous. … God does not just bring things into existence and then take a hands off approach to them. … God continually supports things in existence, moment to moment, throughout the entirety of their careers on the stage of reality.”60 Alston: “God is the ultimate source of the being of everything other than himself: everything other than God exists only because of the divine creative activity. It is not simply that God initially brings each creature into existence; God’s creative or sustaining activity is continually required to keep the creature in being.”61 Quinn: “I now proceed to state my theory of creation and conservation. It consists of this simple axiom: (A) Necessarily, for all x and t, if x exists at t, God willing that x exists at t brings about x existing at t.”62

Such affirmations of divine conservation represent a broad theistic consensus, even down to our own day. But why think the case for conservation stands or falls with the case for concurrence? In order to appreciate Suarez’s argument, we need to distinguish two senses in which God might be said to conserve, sustain, or hold something in existence: immediate conservation and remote conservation.63

Universal Cause and Cause of Human Acts

27

Immediate Conservation: God immediately conserves x if and only if God immediately causes x to exist throughout the duration of x’s existence, at each moment that x exists. Remote Conservation: God remotely conserves x if and only if God does not immediately conserve x, but x is immediately caused to exist at each moment of its existence, and whatever immediately causes x’s existence is, at the moment it does so, either immediately caused to exist by God, or caused to exist by a series of simultaneously operating causes, whose first cause is God.

The differences between immediate and remote conservation are significant. If x must be immediately conserved by God, then God is intimately related to x as a direct cause of its being. By contrast, if x must be only remotely conserved by God, then even though God’s directly causing some other entity is a necessary condition of x’s existing, God doesn’t directly cause x itself. To mark another difference, if x must be immediately conserved by God, then, assuming God is free in what he does, God has great control over whether x continues to exist at any moment. This control is noticeably less impressive if x must be only remotely conserved by God. In that case, the degree of God’s control over whether x continues to exist will depend on whether the intervening causes between God and x bring about their effects necessarily. Only if they do will God be able to guarantee that x continues to exist by bringing about the effect God directly causes. Moreover, with respect to the question of how God might see to it that x ceases to exist, if x must be either immediately or remotely conserved by God, then God can annihilate x, seeing to it that x ceases to exist instantaneously simply by refraining from his direct causal act, with the consequence that x instantly goes out of being.64 Notice, however, that if God’s conservation is remote, his annihilating x will necessarily involve collateral damage; it will require him also to annihilate his immediate effect along with any entities that intervene between that effect and x. Should a remotely conserving God wish to avoid such collateral damage, he would have to refrain from annihilating. In that case, seeing to it that x ceases to exist would require him to create some new entity, or influence some presently existing one, to destroy x or to impede that which immediately causes x. Given only remote conservation, then, God’s seeing to it that x ceases to exist is considerably less elegant and more costly than it would be given immediate conservation. Let us call those who affirm conservation but reject concurrence mere conservationists.65 Such persons typically hold that creaturely substances must be conserved by God, but that creaturely operations need not be caused by God. Whether this claim is plausible, however, depends on what type of conservation is held to be necessary for substances. If a mere conservationist claims, as suggested from some of the passages quoted above, that creaturely substances require God’s immediate conservation, then he holds that at each moment of a creaturely substance’s existence, God directly causes it to exist.66 But, then, to press Suarez’s argument, it is hard to see what feature belongs to creaturely substances, but not to creaturely operations, such that the former, but not the latter, require that God directly cause them. Granted, substances and operations belong to different ontological categories. But, surely, the

28

Free Will and God’s Universal Causality: The Dual Sources Account

answer to the question “Why must creaturely substances be immediately caused by God?” is not “Because they are substances.” The answer is, rather, something like, “Because they are contingent” or “Because they do not exist in virtue of what they are,” or “Because they are beings by participation.” But, of course, these answers apply to creaturely operations as well as to substances. In short, creaturely substances and creaturely operations are on par, it would seem, with respect to any feature that could reasonably be said to require God’s immediately causing them.67 In addition to being contingent, creaturely operations are clearly distinct in being from, that is, not identical to, the substances whose operations they are. If this weren’t the case, then concurrence would simply follow from immediate conservation, since for God to cause a substance would also be to cause its operations. Still, a mere conservationist who favors immediate conservation for substances might reply to Suarez’s argument by insisting that a necessary condition for needing to be directly caused by God is that a thing be a substance. After all, substances are ontologically basic in a way that accidents or features of substances are not. While the latter exist only by belonging to or inhering in substances, substances exist of themselves. They do not inhere in anything, but rather provide the ultimate subjects in which accidents inhere. Perhaps, then, substances need to be directly caused by God, but accidents, since they inhere in substances that are caused by God, do not. Let us assume, for the sake of argument, that creaturely operations are or consist in accidents or features of substances. Given this assumption, does the distinction between substance and accident explain why creaturely substances need God as an immediate cause yet creaturely operations do not? I don’t think so. The fact that accidents only exist by inhering in substances while substances don’t inhere in anything has, as far as I can tell, no implications for the kind of causal influence substances and accidents require in order to exist. Furthermore, the proposed strategy arguably undermines any metaphysical basis for insisting on God’s immediate conservation of substances. For if an operation is sufficiently supported in existence by inhering in a substance that is immediately caused by God, why shouldn’t a substance be sufficiently supported in existence by being directly caused by another creaturely substance that is immediately caused by God? I am not claiming that there are no differences between the inherence relation and the causal relation. Nevertheless, it is hard to see why God’s mediately conserving creaturely substances by remote conservation won’t suffice once it is admitted that God holds creaturely operations in being through the mediation of the inherence relation. The two cases are more alike than different. In both, a contingent entity depends on God, not directly, as a cause, but only because of its dependence relation to something else that does directly depend on God. Although I see no strict contradiction in denying concurrence while affirming God’s immediate conservation of substances, the affirmation does appear unwarranted and incongruous in light of the denial. There is no incongruity in denying concurrence while insisting on only remote conservation for substances, but the denial of concurrence by those who favor God’s immediate conservation of substances seems arbitrary. Proponents of such a position may respond that difficulties with the claim that God causes creaturely operations warrant rejecting concurrence, but that immediate conservation of substances should nevertheless be retained, since it can be retained

Universal Cause and Cause of Human Acts

29

without contradiction, has tradition on its side, and attributes to God a more impressive relationship to creation than does remote conservation. Notice that this justification of retaining immediate conservation for substances even while rejecting concurrence depends on the contention that God’s causing creaturely operations is beset by insuperable difficulties. Yet, if the main arguments of this book succeed, and those difficulties can be overcome, then retaining immediate conservation of substances while rejecting concurrence seems as unwarranted and incongruous as before. In short, if Dual Sources proves successful, the proponent of immediate conservation for substances will have in the desire for doctrinal harmony a significant motive for embracing concurrence, with its claim that God directly causes creaturely actions.

2.5 A Thomistic Argument from Participation As noted in the previous section, Suarez’s reason for thinking that both creaturely substances and creaturely operations require God as an immediate cause is that both are “beings through participation.” According to Suarez, “Only the First Cause is esseitself-through-its-essence; but every other esse is a participation in that esse and so by its intrinsic nature requires the influence of esse-itself-through-its-essence in order to exist.”68 Thus, Suarez argues against those who deny concurrence by an implicit modus tollens: “If God does not have an immediate influence on every action of a creature, then the created action itself does not require God’s influence per se and essentially in order to exist, even though it, too, is a certain participation in being.”69 The claim that all being other than God is being by participation, and therefore that it must be caused by God, who is being through his own essence, figures prominently in Aquinas’s arguments for DUC. In this section, I will present the “Participation Argument” as it is found in Aquinas, though, as we have seen, Thomas is not alone among scholastic authors in endorsing an argument of this sort.70 One reason for treating this argument last (and treating Aquinas after Suarez) is that it is easily the most difficult and metaphysically ambitious argument we will consider in this chapter. Given constraints of space, it will not be possible to evaluate the argument or attempt to defend it against possible objections. Nor will I be able fully to substantiate my presentation of the argument as the correct interpretation of Aquinas. Still, it would be remiss in a book such as this to ignore what is arguably Aquinas’s most significant philosophical argument for DUC, an argument that Suarez also found appealing. Thus, I here attempt, modestly, to explain the basic contours of the argument, indicating largely in notes the texts that I believe support my interpretation. I am cognizant that the success of the argument depends on controversial premises in fundamental metaphysics that it is beyond the scope of the present work to assess. We can begin with Aquinas’s statement of the argument at Summa theologiae 1.44.1 (for a parallel passage see SCG 3.66.7): It must be said that every being in any way existing is from God. For whatever is found in anything by participation must be caused in it by that to which it belongs essentially, as iron becomes ignited by fire. Now it has been shown above (Q. 3, A. 4)

30

Free Will and God’s Universal Causality: The Dual Sources Account when treating of divine simplicity that God is essentially self-subsisting Being; and also it was shown (Q. 11, AA. 3, 4) that subsisting being must be one; as, if whiteness were self-subsisting, it would be one, since whiteness is multiplied by its recipients. Therefore all beings apart from God are not their own being, but are beings by participation. Therefore it must be that all things which are diversified by the diverse participation of being, so as to be more or less perfect, are caused by one First Being, Who possesses being most perfectly.

The argument can be reconstructed as follows: 1. Whatever belongs to something by participation must be (immediately, efficiently) caused in it by that to which it belongs essentially. 2. Being belongs to God essentially and to everything else that exists by participation. 3. Therefore, the being of everything that exists other than God must be (immediately, efficiently) caused by God. I take it that Aquinas means to establish in this passage that all being distinct from God is immediately, efficiently caused by him.71 That creaturely operations are immediately, efficiently caused by God would simply follow.72 In order to understand the Participation Argument, we must first understand the key terms “by participation” and “essentially.” These terms designate mutually exclusive ways in which a form, actuality, or perfection can relate to a subject. To have a perfection “by participation” is for a subject to have the perfection without being identical to it. To have a perfection “essentially” is for a subject to be simply identical to the perfection in question. For a subject to be identical to a perfection is for it to be that perfection subsisting on its own, such that there is no distinction between the subject and the perfection or between the subject and that in virtue of which the perfection belongs to the subject. In such a case, the perfection does not belong to a subject distinct from the perfection; rather, the perfection is its own subsisting subject, a subject unto itself.73 Why Aquinas thinks that there are perfections that subsist as their own subjects will, hopefully, become clearer in the course of considering how he might support premise (1) of the Participation Argument. Unfortunately, to my knowledge, Aquinas nowhere offers a detailed defense of premise (1), nor am I aware of others who have offered such a defense.74 Nevertheless, I think a defense can be constructed in terms Aquinas would accept. The defense is based on claims Aquinas makes, first, about the metaphysics of having by participation and, second, about the metaphysics of causing perfections. Regarding the metaphysics of having by participation, Aquinas thinks that for a subject to have a perfection by participation implies metaphysical composition between two principles, the participated perfection, and that in virtue of which the subject is distinct from the perfection. Within such a composition, Aquinas relates the perfection and the subject (or the perfection and the principle in virtue of which the subject is distinct from the perfection) to one another as act to potency, and he understands

Universal Cause and Cause of Human Acts

31

the perfection to be received, limited, and individuated (or particularized) by the principle of potency with which it is composed.75 For a subject to have a perfection by participation, then, is for the perfection to belong to the subject as the result of a composition of explanatorily prior principles: the perfection as such and that which enters into composition with the perfection as such. The “perfection as such” is the perfection considered in that explanatorily prior moment before it enters into the composition by which it is received, individuated, and particularized. It is through that composition that the perfection belongs by participation to the particular subject in question, on account of that subject’s now having an individual perfection of that type.76 Regarding the metaphysics of causing perfections, Aquinas holds two claims that help explain his support of premise (1). First, he thinks that what has a perfection by participation must have that perfection immediately, efficiently caused in it. That Aquinas affirms premise (1) is by itself evidence that he accepts this claim. Second, Aquinas thinks that the efficient cause of a perfection must somehow contain the perfection that it causes. To deny this principle would, for Aquinas, be tantamount to allowing that something could come from nothing. Yet, Aquinas does not think that a perfection caused must be found in the cause of that perfection in the exact same way or according to the exact same formality. It is possible for a caused perfection to be found in its cause as a higher or more encompassing perfection: Whatever perfection exists in an effect must be found in the effective cause: either in the same formality, if it is a univocal agent—as when man reproduces man: or in a more eminent degree, if it is an equivocal agent—thus in the sun is the likeness of whatever is generated by the sun’s power.77

I note that Aquinas’s allowance here of “equivocal agents,” or causes that contain according to higher formalities the perfections they cause, is not essentially bound up with his particular example of the sun: One could accept the metaphysical point, but reject the particular example.78 We are now in a position to appreciate at least one reason Aquinas, and anyone who agrees with him about the foregoing claims, would accept premise (1) of the Participation Argument. As we have seen, Aquinas thinks that whatever perfection belongs to a thing by participation must be immediately, efficiently caused in that thing. And he also thinks that whatever perfection is efficiently caused must be contained in said cause. These claims by themselves would not give us reason to accept premise (1). But a reason emerges if we add to these points Aquinas’s claim about the metaphysics of having by participation: namely, that to have a perfection by participation is to have it as the result of a composition between the perfection as such and a principle that receives and individuates the perfection as such, so that the perfection is individualized to the particular subject to which it belongs by participation. Given this last claim, a perfection P, as such, is explanatorily prior to any subject’s having P by participation, since the perfection as such is what enters into the composition through which an individual perfection of that type comes to belong to the subject. Why is this significant for a defense of premise (1)?

32

Free Will and God’s Universal Causality: The Dual Sources Account

Given what has been said, any subject that has a perfection P by participation must have P immediately, efficiently caused in it by a cause that contains P. A cause that contains P contains P either essentially or by participation. But a cause that contains P by participation could not, at least by itself, cause P in something else to which P belongs by participation. For, to be the exclusive cause of P in something that has P by participation, a thing would have to be capable of causing P as such, since P as such is that which enters into the composition through which a subject comes to have P by participation. But nothing to which P belongs by participation is capable of causing P as such. Why not? The power in virtue of which an efficient cause brings about an effect is explanatorily prior to that effect. Yet, as we have seen, a thing can cause P—including P as such— only if it possesses P. Thus, in order for a thing to have the power to cause P as such, its possessing P must be explanatorily prior to P as such. But, as we’ve also seen, for nothing that has P by participation is its possessing P explanatorily prior to P as such. On the contrary, for all such things, P as such is explanatorily prior to their possessing P. It follows that nothing to which P belongs by participation is able to cause P as such, and, thus, nothing to which P belongs by participation is, by itself, capable of causing P in something else to which P belongs by participation. And now we can appreciate why Aquinas accepts premise (1) of the Participation Argument. That which has P by participation must have P immediately, efficiently caused in it by something that possesses P and is capable of causing P as such. But nothing that has P by participation can cause P as such. So, whatever has P by participation must have P immediately, efficiently caused in it by that to which P belongs essentially, which is just what premise (1) claims.79 With the defense of premise (1) in place, we can now appreciate why Aquinas thinks there are perfections that subsist as their own subjects. Ultimately, such subsisting perfections are required in order to account for the perfections in things that have their perfections by participation.80 Yet, even though premise (1) requires that whatever has a perfection by participation has that perfection caused in it by that to which the perfection belongs essentially, Aquinas does not think the premise rules out an efficient causal role for that which has P by participation in bringing about P in other things.81 I will not here pause to examine Aquinas’s understanding of that role, but in the following chapter I will argue in a similar spirit that God’s being the direct cause of all entities distinct from himself does not rule out creaturely agency. Now to premise (2) of the Participation Argument, which can be treated much more briefly. Premise (2) states that “being belongs to God essentially and to everything else that exists by participation.” The premise clearly assumes that being or existence is something that belongs to concrete individuals or entities. The assumption is controversial. But, as has been pointed out by a number of its recent defenders, the assumption has much to be said for it and can arguably be vindicated against common objections.82 Since a thing’s existence is that in virtue of which it is actual, existence also counts as a principle of perfection for Aquinas, who understands principles that convey actuality to be perfective.83 Why, then, think that being belongs to God essentially and to everything else by participation? Well, any subject that has existence essentially exists simply in virtue of

Universal Cause and Cause of Human Acts

33

itself; its existing does not call for any explanation outside itself. Thus, if, as Aquinas believes, there are subjects that do not exist in virtue of themselves, then there are subjects that exist by participation.84 Yet, given the reasoning we have already seen for premise (1), if there are things that exist by participation, then there is at least one thing that exists essentially and which causes the existence of that which exists by participation. So far, then, we can see why Aquinas holds that some things exist by participation and at least one thing exists essentially. Yet, Aquinas also thinks that for any perfection P, it is not possible for there to be more than one subject that is essentially or subsistent P. For if subjects x and y were both identical to P, then there would be nothing to constitute the one subject as distinct from the other.85 Thus, there could be only one subsistent P and, likewise, only one subsistent existence (or thing that has existence essentially). And, with this, we can finish Aquinas’s rationale for premise (2) of the Participation Argument: Since all that exists either exists essentially or exists by participation; and since there can only be one thing that exists essentially; and since whatever exists by participation is caused to exist; and since it can’t be that God is caused to exist, it follows that being belongs to God essentially and to everything else that exists by participation. Taken together, premises (1) and (2) entail the conclusion of the Participation Argument, namely, that the being of everything other than God must be (immediately, efficiently) caused by God. I trust one can now see why I described this argument as the most difficult and metaphysically ambitious of the arguments for DUC that we have examined in this chapter. While controversial, the chief claims needed to defend the argument are presumably ones with which some contemporary philosophers—certain Aristotelians, Thomists, and Suarezians—would be sympathetic. The argument, then, is of more than antiquarian interest, even though I have not had space to evaluate it. Among the chief aims of this book is to reconcile DUC with claims that many theists wish to affirm but with which DUC appears incompatible. These aims are especially worthwhile to the extent that DUC is a doctrine theists should take seriously. One reason to take it seriously is its firm place within traditional theism, discussed in Section 1.1. The present chapter has considered additional reasons to take it seriously by considering five separate arguments for DUC, an argument from scripture and four philosophical arguments. I suspect that, for many theists, one or more of these arguments will be at least this compelling: It will give the theist reason to accept DUC provided that it can, in fact, be reconciled with other things he wants to hold. In the following chapter, I consider whether DUC can be reconciled with productive creaturely agency.

3

Divine Universal Causality and the Threat of Occasionalism

In the previous chapter, we considered various arguments for the doctrine of divine universal causality (DUC) and for its corollary that God causes creaturely actions. The corollary raises concerns about whether creaturely freedom could coexist alongside DUC. But even apart from concerns about freedom, it might be thought that DUC jeopardizes creaturely agency. For, many have claimed that God’s causing all being apart from himself leaves no room for productive creaturely causes and thus that embracing DUC commits one to a form of occasionalism. The present chapter considers objections, both metaphysical and epistemic, to what I call Non-Occasionalist DUC (NODUC, for short), a view that holds that everything with real existence (other than God) is brought about by God, yet that most of these effects are also brought about by creaturely causes. It concludes that none of these objections successfully shows that embracing DUC forces one to deny creaturely efficacy. The chapter also considers arguments against the possibility of an agent-causal act being caused by God, showing that these arguments likewise fail. It is open to the proponent of Dual Sources to affirm the existence of productive creaturely causes and even to embrace an agent-causal account of free creaturely action.

3.1 Does DUC Render Creaturely Causes Otiose? Even apart from worries about libertarian freedom, it might be thought that DUC yields problems for an account of creaturely agency. We ordinarily assume that creatures act causally, that—to take an example—this fire causes this heat in this water. It follows from DUC, however, that the whole of the heat is immediately caused by God, since the whole of the heat exists and is distinct from God. But if the whole of the heat is caused by God, could the fire also have caused the heat? Can we affirm creaturely or secondary causation alongside DUC? Or does affirming DUC require us to accept occasionalism, the radical and arguably objectionable denial that creatures are efficacious?1 Certainly, scholastic proponents of DUC, such as Aquinas and Suarez, affirmed creaturely efficacy. On their view, creatures bring things about, with the whole of any creaturely effect being brought about by the creature as well as by God. Thus, in

36

Free Will and God’s Universal Causality: The Dual Sources Account

speaking of God and creaturely causes, Suarez maintains, “it is not the case that part of the effect comes from the one cause and part from the other; rather, as St. Thomas notes … the whole effect comes from each.”2 By contrast, many contemporary philosophers have agreed with the seventeenthcentury thinker Malebranche that DUC rules out genuine creaturely efficacy.3 Timothy Miller, for example, rightly notes that if necessarily God causes all creaturely individuals and all their properties, then no creaturely or secondary cause can bring about an effect without God’s concurrence in bringing about that effect. Yet, Miller doubts that this account leaves room for secondary causes to make genuine contributions: If their contributions really are genuine, it would seem secondary causes should be able to accomplish something without assistance or concurrence.4

According to Philip Quinn: If divine volition does more than bringing about contingent existence [of individuals] and also brings about events in nature [that is, individuals having properties at times], … then no event-causes in nature can be causes of such events.5

William Vallicella likewise argues that God’s bringing about an effect rules out a creature’s producing that effect: If God is the productive cause of the change in the stone from cold to non-cold, then it is entirely superfluous to suppose that the sun or the sun’s shining on the stone plays a productive role in this change. What would remain for the sun to do?6

Hugh McCann and Jonathan Kvanvig reason similarly: Both in the beginning and thereafter, all that obtains does so as a direct consequence of God’s will. But then the entire state of the universe at every instant, all that is and all that occurs, is directly owing to the creative activity of God. It appears to follow that there is no role left for secondary causes; God has, as it were, saturated the world with His own causal power so as to make all other causes otiose.7

To the proposal that a single effect might be brought about by both God and a creature, McCann and Kvanvig respond: Equally unsatisfactory would be to attempt to solve the problem simply by calling for wholesale overdetermination of natural events, making event causation as well as the direct action of God responsible for their existence. In effect, this is not a solution but a restatement of the problem. There is no reason for God to create a world in which natural causes bring events to pass if in fact His very creation of that world includes bringing about those events Himself.8

DUC and the Threat of Occasionalism

37

Vallicella, McCann, and Kvanvig allow that certain—what we might call neo-Humean— accounts of secondary causation are consistent with God’s causing all that exists apart from himself. For example, secondary causation can be affirmed if a secondary causal relation holds between events simply in virtue of their following a pattern that events of their type regularly follow, or in virtue of relations of counterfactual dependence, or some such.9 What is ruled out by these authors is that any secondary cause is genuinely productive of its effect; that it brings about its effect; that the heat really owes its existence to the fire.10 But is the choice between DUC and productive creaturely causation really necessary or can one affirm the position I call Non-Occasionalist DUC (NODUC), a view according to which everything with real existence (other than God) is brought about by God, yet most of these effects are also brought about by creaturely causes?11 It is not always clear what is being charged by those who say that if the heat were caused by God and also by the fire, then one of these causes would be otiose. On a stronger, metaphysical reading, the charge could be understood as claiming that it is literally impossible for the whole of the heat to be brought about by God and also by the fire. On a weaker, epistemic reading, the charge could be understood as denying that there could be warrant for positing more than one productive cause of the heat. Doesn’t positing a second cause where one cause will do violate canons of parsimony?12 In Section 3.2, I discuss in more detail NODUC’s understanding of the relationship between God and creaturely causes, an understanding that I will assume in developing the Dual Sources account. In Sections 3.3 and 3.4, I consider versions of both the metaphysical and epistemic objections to NODUC, and argue that none succeeds. In Section 3.5, I examine arguments for the claim that an agent-causal act (or event), central to certain libertarian accounts of agency, can’t be caused, showing that these arguments also fail. Our discussion over these sections will discover no good reason to think that embracing DUC requires giving up either productive creaturely efficacy or agent-causal understandings of free creaturely action.

3.2 God and Creaturely Causes: The Claims of Non-Occasionalist DUC According to DUC, necessarily, for any entity distinct from God, God directly causes that entity to exist at any time it exists.13 NODUC affirms DUC and also maintains that creaturely or secondary causes bring about effects in the world. In what follows I will, in accordance with scholastic tradition, speak of efficient causes as substances rather than events. Substances are enduring subjects with causal powers. The exercise of causal power by a substance is the bringing about of an effect, a causal act.14 How does NODUC understand the relationships between God, creaturely causes, the effects of creaturely causes, and creaturely causings (or causal acts)? Before answering the question, I note that it is highly plausible to expect that these relationships are sui generis. It should not, then, count against NODUC if we cannot find an example of two creaturely causes that are related to each other in exactly the same way God is related to creaturely causes. The best we may be able to do is to draw various imperfect analogies or comparisons to more familiar creaturely relationships. But the fact that

38

Free Will and God’s Universal Causality: The Dual Sources Account

God’s relationship to creaturely causes may be unique does nothing by itself to show that NODUC is incoherent or unintelligible. If we continue to use the fire’s causing the heat as our example of creaturely causation,15 NODUC entails that the heat, though brought about by the fire, is also simultaneously, directly brought about by God. Indeed, according to NODUC, all that exists in the state of affairs, “the fire’s causing the heat in the water,” is directly brought about by God. Thus, God directly produces the fire, the water, the heat, and whatever other entities (if there be any) that constitute the “fire’s causing the heat.” In some instances, a creature may cause only its effect’s beginning to exist; in other instances, a creature may cause its effect for a time, but the effect outlasts its cause. An example of the former might be parents, who cause their offspring’s beginning to exist, but not its continuing existence subsequent to the beginning (however they may support it). An example of the latter might be a fire that gives heat to the water for a time and is then extinguished, though the heat remains. The accuracy of these examples could be challenged. The important point, however, is that, given NODUC, God brings about the existence of the effect over the entire course of its existence, even if the creaturely cause brings it about only at the very beginning or for some portion of its career. Since the creature, its effect, and its bringing about its effect are all included within the object that God brings about, NODUC understands secondary causes, although genuine, to be subordinate to God, the Primary Cause. Aquinas sometimes speaks of secondary causes as instruments by means of which God brings about certain effects, an analogy that NODUC can happily embrace, provided it is not misunderstood.16 The fire can be spoken of as the instrument or means by which God brings about the heat in the water, since God chooses to bring about the heat in this manner, as opposed, say, to bringing about the heat with a different secondary cause or with no secondary cause at all. Yet, NODUC precludes understanding the analogy to imply that the creaturely cause is a more immediate cause of its effect than is God. If I break a window by means of throwing a baseball, the baseball is my instrument, and it appears to bear a more immediate relationship to the window’s shattering than I do. Given NODUC, by contrast, God directly causes the heat every bit as much as “his instrument” the fire does. A helpful analogy may be found in the relationship between authors and their stories. All the parts of a story are with equal immediacy dependent on the story’s author, even though the author brings about some things within the story by means of others. Tolkien brought it about that Smaug died by means of an arrow shot from Bard’s bow, but the dragon and the archer, the bow and the arrow, and everything else in the story are with the same immediacy dependent on Tolkien. In virtue of their properties, instruments have causal powers, which they exercise, even though they exercise those powers under the direction of the principal agent that uses the instruments. If I wield a saw to cut wood, although the saw’s power to cut is put to work by me, it is the fact that the saw has such properties and power that make it a suitable instrument for my purpose. A hammer would be of no use. Analogously, creaturely substances have diverse causal powers, making them suitable instruments for bringing about different sorts of effects under the direction of the Primary Cause. The fire has the power to produce heat, and its production of the heat is an exercise of

DUC and the Threat of Occasionalism

39

that power. It is not possible, however, for the fire to cause the heat unless God brings about not only the fire and its power but also the heat and the fire’s causal act. The exercise of the fire’s power is, therefore, conditional on God’s concurrence in causing both the heat and the fire’s producing the heat.17 Since God and the fire both bring about the heat, NODUC affirms that God and the fire “co-operate” in causing the heat. I use “co-operate,” with the hyphen, to signify that there are two causes, God and the fire, each of which operates, bringing about the whole of the heat. When a substance brings about an effect, the whole of the effect, not just a part of it, is referred to that substance as its cause. Thus, whatever effects are brought about by creaturely substances are brought about by those substances in whole, not just in part. And since those effects are also brought about by God, they are brought about by him in whole, too.18 “Co-operation” is to be distinguished from “cooperation.” I use “cooperation,” without the hyphen, to signify a situation where two (or more) causes work together to bring about an effect, each contributing only a portion of the effect. Elizabeth and Cecilia, each cleaning a portion of the room, cooperate in bringing it about that the room is clean. Where NODUC will affirm that God and the fire each brought about the heat, it would strictly be incorrect to say that either Elizabeth or Cecilia brought it about that the room is clean. For X to bring about Y implies that the whole of Y is referred to X as its cause. But the whole of the room’s being clean cannot be referred to either Elizabeth or Cecilia, since each brought about only a portion of the room’s being clean. As Freddoso interprets them, scholastic concurrentists, though agreeing that the whole of the effect is brought about by the creature and also by God, nevertheless also held that “the contributions of God and the secondary agent are complementary.”19 Because the scholastics were committed to each of the causes’ bringing about the whole of the effect, they could not understand these complementary contributions by conceiving of the effect “as itself a conjunction of two effects, one which is brought about directly and independently by God and the other of which is brought about directly and independently by the secondary cause.”20 The only alternative was to claim, instead, “that certain features or aspects of the unitary effect are traceable primarily to God and that certain other features of the unitary effect are traceable primarily to the secondary agents.”21 More specifically, according to Freddoso, they maintained that, in an instance of concurrence, one and the same effect has its existence from God, the “universal” or “general” cause, while having its determinate species from the creaturely agent, the “particular” cause. Thus, God allows the creaturely agent to play the role of determining the “what” of the effect, while God accounts for the fact that the effect is something rather than nothing.22 As Freddoso describes the view, “God acts as a universal cause whose proper effect is esse-as-such, while the secondary cause, so to speak, directs God’s universal agency toward its own proper effect, that is, toward a particular effect to which its natural powers are ordered in the relevant concrete circumstances.”23 Although I am hesitant to reject the view just described, neither do I affirm it or include it as a part of NODUC. Tracing some of the features of the effect to God, and others to the creaturely agent, would likely provide a way of avoiding the objection that

40

Free Will and God’s Universal Causality: The Dual Sources Account

one of the causes is otiose. But I think the objection fails even without this move. And I am not convinced that the move is really consistent with the claim that the creaturely effect is wholly from God, and also wholly from the creature, a claim, as we’ve seen, that is explicitly endorsed by Aquinas and Suarez.24 Indeed, attributing some features to God and some to the creaturely agent sounds very much like dividing the effect into parts, with each cause contributing only a part of the whole. Moreover, it is not obvious that the move is really consistent with DUC. After all, the real features of a thing would seem to be entities of some sort. It would, thus, violate DUC to attribute some of these to the creature but not to God. In particular, for scholastics, the features of a thing are due to its substantial and accidental forms, and as Aquinas tells us in a passage quoted in Section 1.1, “whatever is the cause of things considered as beings [namely, God], must be the cause of things, not only according as they are such by accidental forms, nor according as they are these by substantial forms, but also according to all that belongs to their being at all in any way.”25 Even if we can make sense of God’s contributing indeterminate esse, which is given determination exclusively by the secondary agent, if God is not the cause of the forms in virtue of which the effect has its determinate features, then, contrary to DUC, he is not the cause of all entities distinct from himself.26 I leave open the possibility that these objections admit of satisfactory answers. Since, however, I do not think adopting the move is necessary in responding to the charge that, given DUC, either God or the creaturely cause is otiose, I do not include the move as a part of NODUC, and I leave a possible defense of the move to others.27 Finally, although on NODUC the whole of whatever a creaturely substance brings about is referred to that substance as its cause, NODUC denies that any creature is a “complete” cause of what it brings about. As I will use the term, a “complete cause” of an effect E is one such that, by knowing its causal activity, one would know all the causal activity involved in the production of E. No creaturely substance is a complete cause of its effect, since every creaturely effect is (necessarily) brought about also by God and since knowing the causal activity of the creature is not enough (at least by itself) to know God’s activity in bringing about the effect.28 God, on the other hand, is a complete cause of absolutely every entity besides himself. The reason for this is that everything other than God is brought about by God either without the cooperation of a secondary cause or with the co-operation of a secondary cause. If it is brought about without the co-operation of a secondary cause, then in knowing God’s causal activity, one would know all the causal activity involved in the production of the thing. If it is brought about with the co-operation of a secondary cause, then in knowing God’s causal activity, one would also know all the causal activity involved in the thing’s production. That’s because, in such a case, the object of God’s causal activity includes not only the thing in question but also the secondary cause that produced it and the secondary cause’s act of causing it. Since the secondary cause’s activity is part of what God is bringing about in God’s activity, to know God’s activity would be to know also the activity of the secondary cause. Given NODUC, to know God’s causal activity in full would thus be to know the entire causal history of the world he has created.29

DUC and the Threat of Occasionalism

41

Having presented NODUC’s understanding of the relationship between God and creaturely causes, I now return to the question of whether NODUC renders the divine or creaturely cause otiose. Let us begin with the metaphysical objection.

3.3 Non-Occasionalist DUC: The Metaphysical Objection The metaphysical objection holds that it is literally impossible for the heat to be brought about by God and also by the fire. Yet there doesn’t appear to be anything prima facie contradictory in the state of affairs: The fire brings about the heat in the water, and God brings about whatever exists in the fire’s bringing about the heat in the water.

Since embedded in this state of affairs is the heat’s being brought about by God and also by the fire, the metaphysical objection needs some good reason for thinking that the heat can’t be brought about by both.30 Of course, one way to show that a conjunction is impossible is to show that one of the conjuncts is impossible. If one could show that it is impossible for fire to bring about heat in water, or for God to do so, then one would have shown it impossible that “the heat is brought about by the fire and the heat is brought about by God.” In the following, I will assume that both our conjuncts are independently possible, and I will limit myself to considering reasons for thinking impossibility results from their conjunction or from their conjunction coupled with DUC’s claim that the heat can’t exist unless caused by God. I will consider three such reasons. Ted Sider considers the following picture that might underlie a first reason for thinking that an effect’s having two productive causes is metaphysically impossible: Causation is a kind of fluid divided among the potential causes of an effect. If one potential cause acts to produce an effect, that fluid is used up, and no other potential cause can act.31

To develop this picture, one might suppose that causes are to bringing about an effect as jugs of water are to filling a glass. Just as a glass is limited in the quantity of water it can receive, so for any effect there is a finite allotment of bringing about of which it can be the object. Various jugs may each contribute a portion of the glass’s volume, but it is not possible, supposing a single jug supplies the whole of the volume, for another jug to contribute any water. Similarly, if, as proponents of NODUC maintain, God (or the fire) brings about the whole of the heat, then the causal allotment for the heat has been used up, so that it is simply impossible for the fire (or God) also to bring about the heat. Something like the foregoing picture seems suggested by McCann and Kvanvig’s worry that if God brings about the entire state of the universe at every instant, then “God has, as it were, saturated the world with His own causal power so as to make all other causes otiose.”32 Call this the saturation objection. It is, of course, true that supposing a single jug supplies the whole of a glass’s volume, it is not possible

42

Free Will and God’s Universal Causality: The Dual Sources Account

for another jug to have contributed any water—for the volume is a finite quantity and, by the very nature of the case, the water contributed by one jug is a quantity distinct from that contributed by another. If we had reason to think that all bringing about were likewise the supplying of a finite quantity and such that every instance of bringing about conferred a quantity distinct from every other instance, then an argument for the impossibility of the heat’s being wholly brought about by both God and the fire could be motivated. But I know of no reason to accept this model as a characterization of “bringing about” as such and simply to assume it would beg the question against NODUC. We have as yet, then, no reason to think it impossible for the very heat that is wholly brought about by the fire also to be brought about by God. Miller presents a second reason for thinking it impossible when arguing against what he calls “strong concurrentism” for holding what he maintains are two conflicting claims33: SC1  Secondary causes make genuine, non-superfluous causal contributions. SC2 Secondary causes can accomplish nothing at all without God’s specific concurrence. NODUC embraces both of these claims: SC1 because it affirms productive secondary causes and SC2 because, since no effect of a secondary cause can exist unless God causes that effect, it is not possible for a secondary cause to bring about an effect without God’s concurrently bringing that effect about. If Miller shows that SC1 and SC2 are incompatible, then he will have shown it impossible for the fire to bring about the heat if the heat can’t exist unless brought about by God. This would not show that it is impossible for a single effect to be brought about by two distinct causes, but it would show that NODUC is impossible.34 Miller’s argument for the incompatibility of SC1 and SC2 proceeds by considering three general ways in which one causal agent can require the assistance or concurrence of another, and arguing that none of these work for the proponent of strong concurrence. Call this the assistance objection. A vulnerability of Miller’s approach is that its success depends on the three ways he proposes exhausting the possibilities.35 One way he neglects to consider is the following: Given an agent’s power and the right antecedent conditions, it is possible for that agent to bring about a certain kind of effect provided that the effect is also simultaneously brought about by another agent on which the effect necessarily causally depends for its existence.

Of course, this is just the sort of concurrence that NODUC says is required from God in order for secondary causes to bring about their effects. The kind of causal power that secondary causes enjoy on this view is real, but its exercise is conditional; it is not possible for the power to be exercised without God’s concurrence, but with God’s concurrence, it is exercised, such that secondary causes truly bring things about. Unless the neglected way is impossible, SC1 and SC2 are compatible, since the neglected

DUC and the Threat of Occasionalism

43

way describes how a secondary cause might genuinely bring about an effect, while requiring God’s concurrence in order to do so. Miller does nothing to show the impossibility of this neglected way in which one causal agent may require the assistance or concurrence of another, and thus he does not show the incompatibility of SC1 and SC2. The closest he comes to an argument for its impossibility are some remarks he makes against his third proposed way in which one agent may require another’s assistance; the third proposed way states: An agent may possess the kinds of causal powers needed to bring about an effect, but not possess them to the requisite degree; e.g. I possess the kinds of causal powers required to lift things, but not to the degree required to lift a piano.36

Against invoking this way as a solution for the strong concurrentist, Miller says: This model of co-operative action is incapable of satisfying both SC1 and SC2. If we suppose that secondary causes really do make genuine, non-superfluous causal contributions, this model offers no reason for thinking that God’s specific concurrence would always be required, contra SC2. That would be like claiming that I have all of the causal powers needed to lift things, but that there could never be anything, no matter how light, that I could lift without assistance; a claim that seems flatly self-contradictory. If secondary causes have genuine causal powers, then they should not always require concurrence in bringing about the aspects of the effect their powers contribute to. Conversely, if they do always require such specific concurrence, it is hard to see why we should think their causal contributions are genuine, contra SC1.37

In evaluating these remarks as a possible refutation of the neglected way, we need to ask what support Miller gives for the key conditional: If secondary causes have genuine causal powers, then they should not always require concurrence in bringing about the aspects of the effect their powers contribute to.

What Miller offers is that if God’s concurrence were always required, [i]t would be like claiming that I have all the causal powers needed to lift things, but that there could never be anything, no matter how light, that I could lift without assistance; a claim that seems flatly self-contradictory.

Indeed, it does seem contradictory to say that I have all the causal powers needed to lift things but could never lift anything without assistance, at least if we stipulate that having all the causal powers needed implies that nothing else’s power is needed in order for me to lift something. Yet, neither NODUC nor Miller’s strong concurrentist claims that secondary causes have all the causal powers needed to bring about their effects. On the contrary, they deny that it is possible that a secondary cause bring about

44

Free Will and God’s Universal Causality: The Dual Sources Account

its effect without a simultaneous exercise of divine power, insisting only that secondary causes, nevertheless, possess genuine causal powers by which they genuinely bring things about.38 To show, then, that it is contradictory for a thing to have all the power needed to produce an effect, even though it is not possible for the thing to produce that effect without the assistance of something else’s power, does nothing to show a contradiction in what NODUC, or the neglected way, or Miller’s strong concurrentism, affirms. Turning directly to Miller’s key conditional, if the “genuine causal powers” referred to in the antecedent are defined so as to preclude the need for God’s concurrence each time they are exercised, then, of course, the consequent of the proposition follows. On the other hand, the consequent does not follow if, as NODUC claims, a creaturely power to bring about is one whose exercise is always conditional on divine concurrence; for, in that case, having a power that it is impossible to exercise without divine concurrence is exactly what we would expect. In short, Miller’s key conditional gains plausibility only if he assumes the very thing he needs to prove, namely, that a genuine causal power does not always require concurrence. Nothing Miller says shows that a power whose exercise is conditional on divine concurrence is impossible, or that it is not a real power, or that when exercised, a secondary cause does not truly bring something about. Miller has not, therefore, shown the neglected way to be impossible, and thus has not shown SC1 and SC2 to be incompatible. Neither, then, has he shown the impossibility of NODUC. We come, at last, to the third and final reason for thinking it metaphysically impossible that the whole of the heat is caused by God and also by the fire. Won’t one of these putative causes be such that it makes no difference to whether the heat exists? And, if it makes no difference, doesn’t that disqualify it from the status of a cause? Call this the makes no difference objection. Clearly, given DUC, we can’t say that God makes no difference to whether the heat exists, for according to DUC the heat can’t be unless God causes it. But we can ask, since God causes the heat, whether the fire makes any difference to whether the heat exists? It may be thought that it doesn’t if the heat would have been brought about by God anyway, even without the fire. This suggestion assumes the following necessary condition for one entity’s being a cause of another: NCC: Necessarily, for any x and any y, if x causes the whole of y at a time t, then there is no other cause, z, of y, which brings about the whole of y at t and which would have done so even in the absence of x’s bringing about the whole of y at t.39

Let’s suppose, for the sake of argument, that DUC is true and that God causes the fire’s causing the heat. Given NCC, whether the fire can be said to cause the heat will depend on whether God would have caused the heat at the same time, even had he not caused the fire’s causing the heat at that time. Presumably, if God had not caused the fire’s causing the heat, he would have been free in whether or not to cause the heat by himself. Thus, given NCC, the fire won’t be a cause of the heat if the counterfactual of divine freedom, “Had God not caused the fire’s causing the heat at that instant, God would have caused the heat at that instant anyway,” is true.

DUC and the Threat of Occasionalism

45

Philosophers disagree over whether there are any true counterfactuals of freedom. If there aren’t any, then God’s causing the heat is consistent with the fire’s satisfying NCC; for it won’t be true that God would have brought about the heat, even without the causal operation of the fire. On the other hand, suppose there are true counterfactuals of divine freedom. On this supposition, and given NODUC, it is certainly possible that there are cases in which God brings about a secondary cause’s bringing about some effect, where God would not have brought about that effect by himself, without bringing about the co-operation of the secondary cause. But, then, it is possible both that God bring about an effect and that a secondary cause of that effect satisfy NCC. So, even if we accept NCC, the objection won’t have shown it metaphysically impossible for an effect to be caused by God and also to be caused by a secondary cause. At most, we would have reason to think that putative secondary causes aren’t real causes in just those cases (whichever they are) where God would have brought about the effect in question even without the causal co-operation of the creature. I think, however, that NCC should be rejected. For it seems that whether a putative cause is a real cause does not depend on whether its putative effect has some other cause that would have brought about the effect independently, but rather on whether the putative cause actually operates to produce the effect. In the present context, a passage from William Alston is suggestive: God could perfectly well bring it about that the cow is sustained without using grass or any other created agency. However, if God chooses to nourish the cow by the ingestion of grass, the grass must do its thing metabolically if God is to do it that way.40

Were God to nourish the cow without the grass or to produce the heat without the fire, what would be missing are the very causal actions of the grass and the fire. There is, then, a difference between a world in which the fire causes the heat and any world in which it doesn’t. Even if the difference is not always in whether the heat exists, there will always be the difference that in one world, but not the other, the fire brings about the heat.41 As we will see below, theists as philosophically diverse as Aquinas and van Inwagen think that a world with productive creatures is superior to an occasionalist world. If God, likewise judging it good that a creature be efficacious, gives that creature causal power and brings about its act of producing an effect, then that act, and the creature’s status as cause, is hardly negligible.42 x is a real cause of y if x brings about y; and, contrary to NCC, its bringing about y does not depend on there being no other cause z of y that brings about y, and that would have brought about y, even had x not done so.43

3.4 Non-Occasionalist DUC: The Epistemic Objection I have considered three arguments for the metaphysical impossibility of NODUC, finding none of them compelling. Not knowing of other arguments for the same conclusion, let us turn to the weaker, epistemic objection, on which the charge that either God or the fire is superfluous in bringing about the heat is understood as

46

Free Will and God’s Universal Causality: The Dual Sources Account

accusing the proponent of such a view of multiplying causes without good reason, contrary to canons of parsimony. But the obvious response is that the proponent of the view has good reason to affirm both causes just in case he has good reason to affirm DUC and still also good reason to affirm that creatures bring things about. Indeed, since reasons for thinking something actual are also reasons for thinking it possible, to the extent that he has reasons to affirm both causes, he has positive reason for thinking NODUC possible. Because the response to the epistemic objection turns on whether there are, in fact, good reasons to affirm both DUC and productive creaturely causes, one can’t ultimately evaluate the charge without examining very carefully the reasons that can be given for and against both these substantive metaphysical positions. That, of course, would take us beyond the scope of the current project. Still, even a brief survey of some of the reasons that might be given in favor of the position suggests it is doubtful that a person who affirms that a particular effect is brought about by either of God or a creature can’t be warranted in affirming that it is brought about by the other, too. To begin with DUC, we saw in Chapter 1 that the doctrine has theistic tradition on its side; and in Chapter 2 we discussed a number of independent reasons for affirming it, including arguments from scripture, from God’s absolute perfection, from certain cosmological arguments, from divine conservation, and from participation. Anyone for whom the doctrine’s traditional status counts in its favor, or who finds plausible one or more of the arguments considered in Chapter 2, has at least a defeasible reason to affirm DUC. Turning to reasons for holding that creatures are productive causes, I doubt our warrant for thinking they are is limited to stumbling upon some item in the world, such as the heat in the water, seeking an explanation of the heat’s existence, and proposing the fire as what accounts for the heat. If that were the full extent of our warrant, then anyone affirming that God brings about the heat would no longer need the fire as an explanation of the heat’s existence and so might no longer be warranted in positing the fire as a cause. Rather, our warrant for thinking things in the world bring about other things may reasonably be thought to include the fact that this is the way it seems in our experience44—not just in our experience of inanimate objects but also in our experience of ourselves. It would take a great deal, I suspect, to convince most people that they never bring it about that a glass is elevated, that a wall has color on it, or that a loved one laughs—or that they don’t have the power to bring about these effects. Of course, beliefs based on the way things seem are defeasible. We must ask: Does learning that God brought about the color on the wall take away your reason to think that you brought about the color when it strongly seems to you that you did? Assuming that NODUC is possible, I think the answer is “no.” In a case where a person has reason to believe A caused E and then comes to learn that B caused E, it is reasonable to continue to believe that A caused E, provided that it is possible that E had both causes and that one’s reasons to believe that A caused E are sufficiently strong (as such reasons seem to be for our beliefs that we bring things about). All things being equal, parsimony may prefer one cause to two, but it doesn’t dictate one cause in a case where two are possible and the evidence for both is strong. Furthermore, affirming a creaturely cause of E and then coming to affirm DUC puts us in a very different position relative to considerations of parsimony than affirming

DUC and the Threat of Occasionalism

47

one creaturely cause of E and then coming to affirm a second, completely independent, creaturely cause of E. To appreciate this difference, consider Jaegwon Kim’s defense of his principle of causal or explanatory exclusion, well known for its role in his influential work in the philosophy of mind. A recent statement of Kim’s exclusion principle reads: No single event can have more than one sufficient cause occurring at any given time—unless it is a genuine case of causal overdetermination.45

In what is perhaps his most extensive defense of the principle, Kim connects explanation to parsimony, highlighting the unifying and simplifying aim of explanation, and arguing that “multiple explanations of a single explanandum are presumptively counterproductive in regard to the goal of simplification and unification.”46 Such multiple explanations place us in an “epistemic predicament, which can be relieved only when we have an explanation of how the explanations are related to one another.”47 The predicament can be overcome, and unity and simplification restored, says Kim, if we come to see either that the multiple explanations or causes in question are not complete or that they are not independent of each other.48 Thus, Kim’s exclusion principle does not rule out any sort of multiple causes or explanations of the same event but “only bans more than one complete and independent explanation of the same event.”49 Kim discusses a number of cases in which two causes, C and C*, of a single event, E, are understood to be not complete, or not independent of one another, with the result that they do not violate Kim’s exclusion principle. His description of one case reads: C and C* are different links in the same causal chain leading, say, from C to C* and then to E. In this case again we do not have two independent causal explanations; the explanans of one, C*, is causally dependent on the explanans of the other, C.50

Although the causal chain Kim describes is unlike NODUC in that the first cause appears to be less immediately related to the effect, E, than the second cause, the case is just like NODUC on the essential point that the second cause is causally dependent on the first. Given NODUC, the fire does not cause the heat independently of God, since the fire, the heat, and the fire’s causing the heat (whatever that consists in) could not exist unless God caused them. Moreover, since God’s effect includes not just the heat but also the fire’s causing the heat, neither can God’s actual causal activity (his bringing about the fire’s causing the heat) be affirmed yet the fire’s causing the heat denied. God and the fire are, thus, not causes operating independently of one another and forming separable causal accounts. To learn that God brought about the heat does not give us a new and separate account of the heat. Rather, it gives us a more complete picture of what accounts for the heat than we had when we thought that the heat was brought about by the fire alone. Because God and the fire do not give us separable causal accounts, parsimony does not urge us to choose between them in the way that it urges us to choose between two complete and independent creaturely causes that are proposed as independent and separable accounts of a single effect.51

48

Free Will and God’s Universal Causality: The Dual Sources Account

Above I emphasized how the witness of experience supports belief in creaturely causes, but I would be remiss not to note that there are also theological reasons to believe in productive secondary causation. Aquinas, for instance, argues that denying such agency would detract from God’s wisdom, power, and goodness in creating, for creatures unable to bring things about are to that extent useless, deprived of the ability to communicate the good they have through action.52 Though not a proponent of DUC,53 van Inwagen echoes some of these same ideas when he maintains that “occasionalism devalues Creation” and insists that “What God has made and now sustains is substance, not shadow.”54 If these authors are correct that a world of productive creatures is superior to an occasionalist world, then we have a straightforward response to McCann and Kvanvig’s contention that “there is no reason for God to create a world in which natural causes bring events to pass if in fact His very creation of that world includes bringing about those events Himself.”55 Finally, to the question “If God causes the heat, what is left for the fire to do?”56 the answer available to the proponent of NODUC is simple: “Cause the heat.” We have found no reason to think the proposal metaphysically impossible. Moreover, it seems doubtful that there can be no justification for affirming both God and the fire as productive causes. In short, on both the metaphysical and epistemic readings, the charge that either God or the fire must be otiose fails.

3.5 Can Agent-causal Acts Be Caused by God? There is a variety of contemporary libertarian accounts of human freedom.57 Among these, agent-causal accounts (of which there is also a variety) seem closest to our prephilosophical understanding of ourselves as agents. They are also the most congenial from the perspective of the scholastic tradition. Agent-causal accounts understand the most fundamental exercise of free human agency to involve a causal relation between an agent and something the agent, or a controlling part of the agent, such as the agent’s will, directly brings about, usually some sort of mental state, such as a state of intention, or some bodily movement. Agent-causation fits well with our commonsense conception of our own agency. If I perform the act of raising my arm, most people will not hesitate to agree that I (or my will) brought it about that my arm rose. If I form an intention to do something, most will not hesitate to agree that I (or my will) brought about my intention to do that thing.58 The scholastic tradition endorses the notion of agent-causation, since, as we noted in Section 3.2, it understands efficient causes generally—and not only free human agents—to be substances endowed with basic powers and dispositions, in virtue of which they bring about their characteristic effects and which bringing about constitutes the acts or operations of these substances. In Persons and Causes, Timothy O’Connor—perhaps the most well-known contemporary defender of agent-causal libertarianism—offers three distinguishable arguments for the claim that agent-causal acts cannot be caused.59 E.J. Lowe concurs with this conclusion, not giving his own arguments but making explicit reference

DUC and the Threat of Occasionalism

49

to the passages in which O’Connor presents his.60 Perhaps surprisingly, O’Connor’s arguments against an agent-causal act’s having a cause are not that such a situation would rule out libertarian freedom. Rather, independently of libertarian concerns, he argues that the idea that an agent’s direct bringing about of an effect is itself brought about (or caused) is incoherent. If O’Connor’s arguments are sound, and agent-causal acts cannot be caused, then DUC is clearly incompatible with agent-causal accounts of human agency. For DUC implies that, necessarily, if they exist, such acts are caused by God, since they are entities distinct from God. In this final section of the chapter, I consider O’Connor’s three arguments and show that none of them establishes that agent-causal acts cannot be caused.61 It is possible that, in developing his arguments, O’Connor was limiting the scope of “cause” to creaturely causes and not even considering the possibility of whether God might cause an agent-causal act. Nevertheless, O’Connor doesn’t indicate that his scope is so limited, and thus a response to his arguments that an agent-causal act cannot be caused is necessary if we want to defend the compatibility of DUC with agent-causal accounts of creaturely freedom and agency. O’Connor’s first argument comes within the context of a discussion of Richard Taylor’s agent-causal account. Taylor had argued in favor of the view that an agentcausal act might be caused.62 O’Connor sets out to show that “Taylor is mistaken.”63 According to O’Connor: When we carefully analyze what is meant by analogous claims in the context of event causation, we see that they do not support the scenario Taylor is envisaging. Despite appearances, such claims are not intended to assert a scenario that involves a direct cause of a metaphysically basic instance of causation.64

O’Connor’s first argument continues by considering examples in the context of event causation where we seem to affirm one event, E1’s, causing the causal sequence between two other events, E2 and E3, but where, properly understood, E2’s causing E3 is not directly caused by E1. O’Connor considers two sorts of examples. In his example of the first sort, “we may sensibly say that my finger’s pressing the button causes the causal sequence, the ringing of the bell’s causing the cat to jump.”65 All we really mean, however, is that my finger’s pressing the button “caused the sequence indirectly, by causing the first element of the sequence, the bell’s ringing.”66 In his example of the second sort, “we may sensibly say that the electrician’s wiring of the doorbell system was a cause of the sequence, the depressing of the button’s causing the bell to ring.”67 But all we mean in this statement is that the electrician’s wiring set up a context in which the depressing of the button caused the bell to ring. We don’t mean that the electrician’s wiring actually “brought about … the button’s depression’s exerting its characteristic influence.”68 O’Connor concludes: Neither of these legitimate ways of speaking of causes of causings within eventcausal contexts supports the idea of a cause of an agent’s causing his own intention. … Hence, we should reject this idea altogether.69

50

Free Will and God’s Universal Causality: The Dual Sources Account

Yet even if one agrees with O’Connor’s analysis of the foregoing cases, this first argument falls short of establishing its intended conclusion. From the fact that two sorts of scenarios, in which we can sensibly speak of one event’s causing a causal sequence between two other events, don’t lend support to the idea that an agent-causal act can be caused, it doesn’t follow that agent-causal acts aren’t caused or that they can’t be caused.70 O’Connor’s second argument comes in the context of discussing an agent-causal account presented by Roderick Chisholm.71 Chisholm had worried that if an agent agent-causes (or makes happen) some event e, we won’t be able to hold the agent responsible for agent-causing e, unless the agent agent-causes (or makes happen) his agent-causing e. In order to safeguard responsibility, Chisholm thus stipulated that agents agent-cause (or make happen) all their agent-causings, even though this stipulation commits Chisholm’s account to an infinite regress or nesting of agentcausings every time an agent agent-causes anything at all. O’Connor finds the regress untenable: “Positing an infinity of such causes of causings is not merely ridiculous on the face of it but also logically vicious.”72 The problem can be avoided, O’Connor thinks, if we simply deny that in order to be in control of his causing e, an agent must cause his causing e. Instead, we should understand that an exercise of agent-causation in agent-causing e is intrinsically an exercise of control, such that no further (or prior) act of the agent is needed in order for the agent to be responsible for his act of causing e.73 O’Connor’s proposed solution to the regress problem seems very sensible. Our question, however, is whether the untenability of Chisholm’s regress shows that there can’t be a cause of an agent-causal act. O’Connor seems to think that it does: “Given that Chisholm’s idea of an infinite nesting is unacceptable, there couldn’t be a cause of a given agent-causal event.”74 But this second argument proves no more successful than the first. What follows from the unacceptability of Chisholm’s regress is not that there couldn’t be a cause of an agent-causal act but only that the claim that leads to the regress must be denied. What must be denied, then, is that agents agent-cause all their agent-causings. And that denial is perfectly consistent with an agent-causing’s having a cause. It is consistent, for instance, with a creature’s agent-causal act being caused by God. O’Connor’s third argument appears as follows: Reflection on the nature of such an [agent-causal] event suggests that there cannot, in the nature of the case, be a cause that produces it. … That something should bring about the causing of an event is absurd: it implies that the cause in the basic causal transaction is in some way deficient, that something further is needed to bring about its causing of its effect.75

It is not clear exactly how this argument is supposed to go, but the conclusion (i) “that there can be no cause of an agent-causing” seems to derive from the more general claim (ii) “that no cause C* can bring about another cause C’s causing an event.” And (ii) apparently derives from claim (iii) “that no cause C can be such that some other cause C* is needed to bring about C’s causing its effect.”76 Like the previous two, this

DUC and the Threat of Occasionalism

51

argument fails to show the impossibility of an agent-causing having a cause. For, O’Connor provides no reason to accept (iii) nor has the discussion in the present and previous sections of this chapter turned up any good reason to accept it. O’Connor has not, then, shown that an agent-causal act can’t be caused, and, thus, he has not given us reason to think that such an act could not be caused by God. Nor have we discovered any good argument for the frequently made claim that God’s causing all that exists apart from himself rules out productive creaturely agency. In short, we have discovered no good reason to think that DUC commits us to a form of occasionalism or that it rules out creaturely agent-causation. A proponent of Dual Sources can embrace both productive creaturely agency and an agent-causal account of free creaturely action.

4

Free Creatures of the Universal Cause

Prevailing opinion holds that if creaturely acts are caused by God, then they cannot be free in the libertarian sense. An implication of this view is that the traditional doctrine of divine universal causality (DUC) rules out libertarian creaturely freedom. But, as this chapter will show, whether God’s causing our actions precludes libertarian freedom depends entirely on one’s account of divine causal agency. On a common way of thinking of God’s agency, there is, indeed, a conflict between DUC and libertarian creaturely freedom. But there is no conflict on what I call “the extrinsic model of divine agency.” Given the extrinsic model, God’s causing a creaturely act does not preclude libertarian freedom by rendering the act determined. Nor does it rule out the creaturely agent’s “ability to do otherwise” or its “ultimate responsibility” as these notions have been understood by representative libertarians. The combination of the extrinsic model and DUC gives us an account on which a free creaturely act has dual sources, God and its creaturely agent, and is genuinely “up to” both in a way that leaves all the conditions for libertarian freedom satisfied. Nor is appeal to the extrinsic model ad hoc. As we will see, the model emerges from central tenets of scholastic theology and fits well with a Thomistic tradition that understands God’s transcendence to be the key to reconciling divine causality and creaturely freedom. Moreover, as I will go on to show in Chapter 5, together with plausible assumptions, the extrinsic model turns out to follow from DUC, and thus to be the model of divine agency the proponent of DUC is committed to anyway, even aside from its usefulness in reconciling DUC with libertarian freedom.

4.1 The Intrinsic/Extrinsic Distinction In preparation for discussing different accounts of divine agency, let me first indicate what I mean by the adjectives “intrinsic” and “extrinsic” in their primary use, which is to divide types of predication and types of entity, relative to a given object. Although the intrinsic–extrinsic distinction is both intuitive to common sense and ubiquitous in philosophy, it has nevertheless proven difficult to analyze in terms that do not presuppose the distinction. Like other philosophers, I assume that even short of a satisfactory analysis, we can establish meanings for the terms adequate for their use.1

54

Free Will and God’s Universal Causality: The Dual Sources Account

Some predications, such as “Cecilia is female,” characterize an object in virtue of its makeup or how it exists in itself. Other predications, like “Cecilia is amusing to Elizabeth,” characterize an object, not in virtue of the way it exists in itself but in virtue of its relations or lack of relations to things that fall outside the object. The former are “intrinsic predications,” the latter “extrinsic predications.”2 The truth of intrinsic predications is grounded solely in what is intrinsic with respect to an object, while the truth of extrinsic predications is grounded, at least in part, in entities that fall outside the object and its makeup. One need not look outside Cecilia to see what grounds the truth of “Cecilia is female,” but to see what grounds the truth of “Cecilia is amusing to Elizabeth,” one must look not just to Cecilia but also to Elizabeth and to the relation that holds between them. “Elizabeth is a mammal” is an intrinsic predication. “Cecilia lives in St. Paul,” an extrinsic predication. “Cecilia has two arms” is an intrinsic predication. “Elizabeth has the same number of arms as Cecilia,” an extrinsic predication. If we assume that beliefs are intrinsic states of believing subjects, then “Cecilia believes that Elizabeth is in the kitchen” is an intrinsic predication. If we assume that knowledge is warranted true belief, then “Elizabeth knows that Cecilia is in the bathroom” is an extrinsic predication, since it is grounded, if true, in things outside of Elizabeth, such as Cecilia’s actually being in the bathroom.3 The truth-value of an intrinsic predication cannot change without the subject of that predication either changing or ceasing to exist. If it ceases to be true that “Cecilia has brown hair,” it will be either because Cecilia has ceased to exist or because she has undergone a change in hair color. The truth-values of extrinsic predications, by contrast, typically can change without the subject’s changing or ceasing to exist. Cecilia may cease being amusing to Elizabeth without any change in Cecilia at all, say, if Elizabeth loses her sense of humor. The term “merely Cambridge change” has entered the philosophical lexicon to name those cases in which a predication comes to be or ceases to be true of an object in virtue of a change in that object’s relations to other things without the object itself changing or ceasing to exist.4 The distinction between real and merely Cambridge change can be used to elucidate the distinction between entities intrinsic with respect to an object and entities extrinsic with respect to an object.5 An entity is intrinsic to (or with respect to) an object if and only if the entity’s ceasing to exist or ceasing to belong to the object would of itself constitute either the object’s ceasing to exist (if the entity were identical to the object or belonged to the object essentially) or the object’s undergoing a real, as opposed to merely Cambridge, change. An entity not intrinsic with respect to an object is extrinsic with respect to that object.6

4.2 Why DUC May Appear to Preclude Libertarian Freedom In Section 1.2, I discussed in detail both the libertarian conception of human freedom and the reasons why so many theists have been attracted to it. Recall from the section that we can distinguish strict and broad accounts of what it is for an act to be free in the libertarian sense. According to the strict account:

Free Creatures of the Universal Cause

55

An act is free in the libertarian sense if and only if its agent performs the act voluntarily and intentionally, and the act is not determined (i.e., there is no factor both prior to and logically sufficient for the act).

According to the broad account: An act is free in the libertarian sense if and only if its agent performs the act voluntarily and intentionally, and either the act is not determined (i.e., there is no factor both prior to and logically sufficient for the act), or the act is determined and the agent’s responsibility for the act derives from the agent’s voluntary and intentional performance of some prior act that was not determined.7

The parenthetical notes in these definitions remind us that, in the sense of “determined” relevant to the definition of “libertarianism,” an event is determined if and only if there is a factor both prior to and logically sufficient for the event, where a factor counts as “prior” if it is causally or explanatorily, even if not temporally, prior to the event. By “voluntary” I mean that the act is something the agent does willingly, as opposed to being coerced. By “intentionally” I mean that the act is done on purpose. I speak of an agent’s “performing” its act, but I do not take the performance to be anything distinct from the act itself. The performance just is the act. Given these definitions, why might someone think that if all our acts are caused by God, then none can be free in the strict or broad libertarian senses? The answer, I suspect, lies in certain common assumptions about God’s agency. One assumption is that God’s causal activity in bringing about a creaturely effect E is causally or explanatorily prior to E.8 For, it is highly plausible that God’s act of causing E is logically sufficient for E—that there is no world in which God is causing E, but E does not exist. But if God’s act of causing E is not only logically sufficient for but also prior to E, then it determines E. And, if God’s act of causing E determines E and all our acts are caused by God, then all our acts are determined, and none can be free in the strict or broad libertarian senses. Another assumption results from modeling divine agency on human agency. Not uncommonly, when human beings perform some bodily act, we think that they make a decision or choice, or form an intention, so to act, and that their bodily movement is causally or explanatorily subsequent to their decision, choice, or intention. It would not be surprising if many thought of God’s causal acts according to a similar model. On this model, God’s causing some creaturely effect E involves his deciding, or choosing, or intending, or decreeing, or willing, to bring about E, and E is causally or explanatorily subsequent to God’s choice or decree: that is, God’s choice (etc.) either causes E, or is part of a cause of E, or in some other way is that in virtue of which God causes E.9 Of course, a human being may choose to perform some bodily act that never eventuates because the choice is impeded or ineffectual. But it is natural to suppose, in light of God’s power, that God’s choice to bring something about cannot be impeded or ineffectual. So, for those assuming a model of the sort under consideration, it is natural to suppose that, necessarily, if God chooses (etc.) to bring about E, then E gets

56

Free Will and God’s Universal Causality: The Dual Sources Account

brought about. And, since we typically think of human choices or intentions as being intrinsic acts or states, it is natural to assume that God’s choices (etc.) are likewise intrinsic. The foregoing model, I suspect, is a fairly popular way of thinking about God’s agency. Let’s call it the popular model (PM, for short). According to PM, whenever a statement of the form “God causes E” is true, the following items will exist: a. God b. E c. God’s choice, decree, or intention to bring about E, which is intrinsic to God, is that in virtue of which God causes E and which would not exist were God not causing E. d. The causal-dependence relation between God and E. PM is a particular example of a broader class of models of God’s agency, what I call “intrinsic models” of divine agency, to be discussed in more detail in Chapter 5. The important point to appreciate now, however, is that, when coupled with DUC, PM clearly rules out a creaturely act’s being free in the libertarian sense and that’s true whether we follow the strict or broad account. For DUC implies that every creaturely act is directly caused by God. And PM implies that for every creaturely effect E that is caused by God, there is a factor—God’s choice or decree (or etc.)—that is both prior to and logically sufficient for E. This factor is prior to E, since it either causes E or is that in virtue of which God causes E.10 This factor is logically sufficient for E, since it is not possible for the factor to occur or obtain without E’s occurring or obtaining. It follows that DUC and PM imply that every creaturely act is determined. But if every creaturely act is determined, then no creaturely act can be free in the libertarian sense: not on the strict account and not on the broad account. Given the likely popularity of PM, as well as the common assumption that God’s act of causing an effect E is causally or explanatorily prior to E, it is no wonder that so many have assumed that DUC rules out libertarian creaturely freedom.

4.3 The Extrinsic Model of Divine Agency But not all models of divine agency, when coupled with DUC, result in our acts being determined—what I call “the extrinsic model” (EM) does not. Recently defended by Jeffrey Brower, Timothy O’Connor, and Alexander Pruss (among others), EM is originally motivated by two doctrines affirmed within scholastic theology: the doctrine of divine simplicity and the doctrine that God is not really, but only rationally, related to creatures.11 Divine simplicity holds that God lacks composition of any sort and that there are no entities intrinsic to, but distinct from, God.12 It follows from divine simplicity that whatever is intrinsic to God is simply identical to God. But if God’s act of causing E or God’s choice to cause E were identical to God, then, contrary to the common scholastic understanding of God’s freedom, God could not have done otherwise than cause, or choose to cause, E.13 One committed to both divine simplicity

Free Creatures of the Universal Cause

57

and God’s freedom to do otherwise, thus, needs a model of God’s agency on which God’s causing or choosing to cause E is extrinsic, not intrinsic, to God.14 Turning to the doctrine that God is not really related to creatures, note that this easily misunderstood claim is not what it sounds like at first. It is not the claim that statements predicating of God a relationship to creatures are not really true or that they are perhaps just metaphorical. Rather, the teaching concerns the ontology of the relations between God and creatures. Put differently, the teaching concerns the ontological implications of statements that predicate of God relationships to creatures. As we will see, the teaching concerns primarily what the ontology of such relations does not include. The substance of the denial that God is really related to creatures together with its implications for an account of divine agency can be appreciated once we get clear on the scholastic distinction between “real” and “rational” relations. Mark Henninger takes Aquinas’s account of the distinction to be representative of the tradition as a whole: [Aquinas] held that a relation R of a to b is real only if a and b are really distinct extra-mental things, and there is a real extra-mental foundation in a for R. Aquinas also held that a relation R of a to b is of reason only if either (i) a and/or b is not real, or (ii) a and b are not really distinct, or (iii) there is no real foundation in a for R.15

Clearly, since both God and his creatures are real and really distinct from one another, God’s relations to creatures cannot satisfy condition (i) or (ii). The claim that God is only rationally (and not really) related to creatures must therefore be understood according to condition (iii): For any relation God has to creatures, there is no real foundation in God for that relation.16 Now, the critical point to note is that the claim that God is only rationally related to creatures applies to every case in which a relationship to creatures is predicated of God—including cases in which God is said to be related to his creatures as knower to known or as cause to effect.17 The consequences for an account of divine causal agency are clear enough. God’s causing or bringing about some effect within creation will not involve any real or intrinsic state or property of God that would not be there were he not causing that effect. What, then, does divine causal agency involve given divine simplicity and the doctrine that God is not really related to creatures? According to O’Connor: His activity entirely consists in a causal relation between Himself … and the dependent, contingent reality. … There’s just (i) an agent with reasons for various possible creations, and (ii) a relation of dependency between that agent and the actual creation, such that the product might have been utterly different, and the agent utterly the same.18

Certainly, this picture of divine agency differs markedly from PM, on which God’s causing E involves an intrinsic feature of God—God’s choice or decree—in virtue of

58

Free Will and God’s Universal Causality: The Dual Sources Account

which God causes E and which feature would not exist were God not causing E. On EM, although God’s causal power is intrinsic to God, the actual exercise of this power in producing an effect does not occur in virtue of PM’s item (c), something intrinsic to God that would not exist were God not bringing about the effect. As the passage quoted from O’Connor indicates, on EM, when a statement of the form “God causes E” is true, only the following need exist: a. God (with God’s reasons) b. E d. The causal-dependence relation between God and E. Of course, were God not causing E, there would be neither E nor the relation between God and E.19 God, however, would be intrinsically the same. Presupposed by EM is the idea that the causal act of an agent—what the agent is doing when it is causing something—may consist solely in the agent’s causing or bringing about some effect, and that an agent’s causing or bringing about some effect may consist in nothing more than the causal-dependence relation between the effect and the agent or in the effect’s standing in a causal-dependence relation to the agent. As we will see in Sections 4.4 and 5.1, this understanding of causal action is not limited to theology. It is the operative understanding of causal action by at least some contemporary agent-causal theorists, and it is the understanding of causal action defended by Aristotle within a purely natural context and adopted by the scholastic tradition that was influenced by him. EM maintains that for any possible creaturely entity, God knows respects in which to cause that entity would be good: either because of that entity’s own goodness; or because the entity has a place in a larger whole that is good; or because the entity in some way manifests or contributes to the manifestation of God’s goodness. For every possible creaturely entity, therefore, God has reasons to bring about that entity. God knows these reasons in knowing himself (or in knowing himself and what he has created), but, in keeping with divine simplicity, these reasons are not entities intrinsic to, but distinct from, God.20 Moreover, these reasons leave God free to refrain from causing any creaturely effect, and refraining is consistent with God’s wisdom and goodness.21 In actually bringing about E, God does so for a reason, and therefore his bringing about E is purposeful and intentional. Because it is purposeful and intentional, we speak truly if we say that “God brings about E intentionally” or “God wills E.” Since God might have refrained from willing E, we can even say that “God chooses E.” Such truths, however, should not be taken to imply that God’s intending E, willing E, or choosing E pick out or involve counterfactually variant intrinsic states of God, as they do, for example, on PM. On EM, God’s willing, choosing, or intending E is not an act distinct from his causing E; it is just God’s causing E for a reason. And God’s causing E is nothing intrinsic to God. Rather, God’s causing or causal act consists entirely in items that are extrinsic.22 What items? It appears there are two possibilities.23 First, God’s act of causing E might consist simply in the causal relation between God and E, the option that O’Connor would seem to favor in the passage above. Second, God’s causal act might

Free Creatures of the Universal Cause

59

consist in the causal relation and E, together. Although either possibility can count as a version of EM, of these options, I prefer the second. The problem with the first is that if a divine causal act is simply a bare causal relation and does not also include the effect God is causing, it looks as if God’s activity, or what God is doing in his various causal acts, is always formally the same. But, surely, what God is doing isn’t always formally the same. For God to cause American Pharoah is for God to do something different than for God to cause Justify; or, if one thinks that God performs a single act in which he causes everything he causes, then for God to perform that act is for God to do something different than he would have done had he brought about different things than he did. A divine causal act is specified by the effect(s) in which it terminates. We should, therefore, recognize the effect and terminus as an essential part of the act. At least, that’s how I will construe EM going forward. In Chapter 5, I will have much more to say in defense of EM. For now, let me close this section with a word about how divine causal acts are counted on this model. The way I have presented EM suggests that there is a distinct divine causal act for every distinct entity to which God stands in a causal relation. But I have also said that when God brings about an effect, he does so for a reason. And, though we are often blind to God’s reasons for bringing about the things he does, we can assume that some of the things God brings about he does for the sake of other things he brings about or for the sake of a whole of which these entities are a part.24 In cases where distinguishable effects of an agent are ordered to a larger whole by the agent’s reasons for bringing about the effects, we can speak of the agent’s bringing about these effects as part of a larger action, which is the bringing about of the whole to which they are ordered. In describing what I did Saturday afternoon, it would be more informative to say that I mowed the lawn than that I cut this patch of grass, and then that patch, and then that other patch, and so on. It is true that I cut each of these patches, and therefore that I performed a number of distinguishable acts. But, given that I cut each patch for the sake of, or as part of, mowing the lawn as a whole, it is also true that I did one thing: mow the lawn. There is no contradiction here, because the claims are true relative to different, equally legitimate, principles for counting actions. One principle says we have a distinct causal act for every distinct thing the agent brings about. Another principle says that whatever an agent does as ordered by reason to a single end or whole is part of a single act.25 In a similar way, it is true on EM that there are as many divine causal acts as there are distinct entities to which God stands in a causal relation; nevertheless, God may cause distinct entities as part of a larger action, if those entities are ordered by the divine reason to other entities or to a whole of which the entities are a part. It is even possible that, in light of divine reason, there is a single divine action of which God’s causing each distinct entity is a part. Suppose that possibility is reality. In my view, it would be correct to say both that God has performed the single act of bringing about of the whole of creation and that God has performed as many divine causal acts as there are entities to which he is causally related. The claims would be consistent because true relative to different principles for counting actions. While recognizing both principles, I typically count divine causal acts according to the principle that recognizes a distinct causal act for every entity that bears a cause–effect

60

Free Will and God’s Universal Causality: The Dual Sources Account

relation to God. The arguments of the book are not affected by this choice, but the choice makes sense, since counting according to the other principle would require making possibly controversial assumptions about God’s reasons for bringing about what he does. According to my typical way of counting, then, a divine causal act on EM simply consists in the effect caused and the causal relation that holds between God and the effect. God’s intending, willing, and choosing the effect consist in nothing over and above God’s act of bringing the effect about. “Intending,” “willing,” and “choosing” simply add to “causing” and “bringing about” the idea that what God brings about he does for a reason, and, in the case of “choosing,” that God could have done otherwise.

4.4 DUC without Determinism Why, then, does adopting EM render consistent God’s causing our acts with their being free in the libertarian sense? Because, on EM, God’s causing a creaturely act A (whether a bodily movement or a choice or act of will) need not introduce any factor that “determines” A—that is, any factor both prior to and logically sufficient for A. Consider the items on the scene when God causes A, assuming EM. ● ● ● ● ●

●

God A God’s reason for causing A The causal-dependence relation between God and A God’s causal act, or causing of A, which consists in A plus the causal relation between God and A. God’s willing or choosing A, which is nothing else than God’s causing A for a reason when God could have done otherwise.

Do any of these items constitute a factor both prior to and logically sufficient for A? Well, as cause of A, God is causally prior to A. But God is not logically sufficient for A: It is possible for God to exist without A. A, by contrast, is logically sufficient for itself. But clearly A is not prior to itself. Thus, neither God nor A is a factor both prior to and logically sufficient for A. God’s reason for causing A is arguably prior to A. But on EM, at least if we assume DUC, God’s reason is not logically sufficient for A. That’s because, on EM, God’s reasons for causing creaturely effects leave him free to refrain from causing those effects. But, given DUC, it is not possible for anything distinct from God to exist without God’s causing it. So, given EM, DUC, and any creaturely effect E that God causes in the actual world, there is a possible world in which God has the same reason to cause E, but in which E does not exist. But, then, God’s reason is not logically sufficient for E. And so neither is God’s reason for causing A logically sufficient for A.26 What then of the causal relation between God and A? And what of God’s act of causing A or God’s willing and choosing A? We need not determine whether the causal relation between God and A is logically sufficient for A, for it is not prior to A; relations are not prior to their relata.27 God’s act of causing A is certainly logically sufficient for

Free Creatures of the Universal Cause

61

A: There is no world in which God causes A, but A does not exist. Yet God’s causal act is just as clearly not prior to A. For that act consists in A together with the causal relation between God and A, and, as we have just seen, neither of these is prior to A. And, since God’s willing or choosing A is just God’s act of bringing about or causing A for a reason, God’s willing or choosing A is not prior to A, either. In short, given EM, none of the items on the scene when God causes A constitutes a factor both prior to and logically sufficient for A. But, then, on EM, God’s causing A does not render A determined. What goes for A goes for any creaturely act. Given EM, God can cause all creaturely acts without rendering any of them determined. Thus, on EM, the only way a creaturely act caused by God would be prevented from being free in the libertarian sense is if God’s causing such an act precluded its creaturely agent’s performing the act voluntarily and intentionally. But I know of no good reason to think that an act’s being caused by God rules out its creaturely agent’s performing the act willingly and on purpose. So, given EM, it seems that all human acts can be caused by God and still satisfy both the strict and broad conditions for being free in the libertarian sense. Yet, one may object, don’t God’s causal acts in some sense explain his effects? If so, doesn’t that mean that God’s causal acts are explanatorily prior to his effects, with the consequence that God’s act of causing a creaturely act A constitutes a factor both prior to and logically sufficient for A, after all? Clearly, if what I have argued above is correct, God’s acts do not explain his effects in the paradigmatic way by causing them; for a cause is prior to its effect, but, as we have seen, God’s act of causing E is not prior to E. Why, then, might one still be tempted to think that God’s causal acts are explanatorily prior to his effects? Perhaps, it will be thought almost a truism that agents cause effects by means of their actions, which suggests that those actions are explanatorily prior to the effects caused. Yet, while it is true that agents sometimes bring about effects by means of actions, it is doubtful that all of an agent’s effects are so brought about. I may cause the ball to fly through the air by means of swinging the bat. Yet, philosophers have often recognized a class of actions—so-called “basic actions”—that consist in an agent’s causing or bringing about an effect, but not by means of anything at all.28 Consider, for instance, a proponent of agent-causation who thinks that there are basic acts of the agent that consist in the agent’s directly causing a bodily movement or, alternatively, an executive state of intention.29 In such cases, there is no prior act with its own object, such as the swinging of the bat, by means of which a further effect, such as the ball’s flying through the air, is brought about. Rather, the basic act is nothing other than the direct bringing about of the effect in question. Should we say that such direct effects are caused by means of the agent’s basic acts? But, since these acts just are the direct causing of these effects, answering this question in the affirmative would be tantamount to saying that the agent causes these effects by means of causing them, a suggestion that, far from giving us a genuine means by which the agent causes the effects, merely reiterates the fact of his causing them. When it comes to basic actions, then, an agent causes an effect, but not by means of anything, including the agent’s causal act. The point, of course, is that on EM God’s causal acts are basic actions. Consequently, if we assume this model, we should think of God’s causal acts as consisting in his causing effects, not

62

Free Will and God’s Universal Causality: The Dual Sources Account

of God as causing effects by means of his causal acts. We should say that God brings about his effects in his causal acts, not by means of his causal acts. Yet, surely, the theist will think that citing the state of affairs God’s causing X or the fact that God caused X explains the existence of X. Surely, the truth of the proposition God caused X explains the truth of the proposition X exists. Surely, if someone asks “Why does X exist?” the answer “Because God caused it” at least partially satisfies. Were the proponent of EM forced to deny these points, would that not reduce his account to absurdity? But the proponent of EM need not deny these points. He can happily agree that to answer “Because God caused X” or to reference the state of affairs God’s causing X or the fact that God caused X explains the existence of X. It does so because, in each case, it identifies a cause of X and to identify a cause is the paradigmatic way of explaining something. Nevertheless, the proponent of EM will insist that the cause identified in these cases is not God’s act of causing X but rather just God. Statements such as “X exists because God caused it” or “God’s causing X explains the existence of X” are ambiguous. What actually accounts for or brings about X’s existence here? Is it the whole state of affairs God’s causing X? Is it God’s act of causing X? Is it simply God? The proponent of EM will say that, of these options, it is only and strictly God that causes or brings about X’s existence. But that’s perfectly consistent with recognizing that we have offered an explanation in referencing the fact that God caused X or the state of affairs of God’s causing X or the truth of the proposition God caused X. Indeed, on EM, it seems that a divine causal act cannot be explanatorily prior to the effect brought about in that act, and this for the same reason that only God, and not God’s causal act, can cause such an effect. For, whatever causes or is explanatorily prior to E’s existence cannot itself presuppose E’s existence. As I am using the term “presuppose,” “x presupposes y” does not mean merely that x is logically sufficient for y and y logically necessary for x, a relationship perfectly consistent with x’s being causally or explanatorily prior to y, as in the case of causes that determine (or necessitate) their effects. Rather, I am using “presuppose” to cover a broad family of relationships that would preclude x’s being prior to y. “x presupposes y,” if x exists in dependence on y, or consists (at least partially) in y, or has y as an essential constituent, or if y is essentially part of what is designated when referring to x, or if for some other reason x cannot exist prior to y or independently of y. While “x presupposes y” precludes x’s existing prior to or independently of y, it leaves open the question of whether y exists prior to x, which question will depend on further details of their relationship.30 Now, on EM, God’s act of causing E clearly presupposes E’s existence, since it partially consists in E. Even on a version of EM on which God’s act of causing E consists in just the causal-dependence relation between God and E, that act will still presuppose the existence of E, for relations presuppose the existence of their relata. Since, on EM, God’s act of causing E presupposes E, that act cannot be explanatorily prior to E and cannot in any sense cause or account for E. Is EM unique in denying that a causal act is explanatorily prior to the effect brought about in that act? Reflecting again on the theory of agent-causation, it should be clear that neither is an agent’s basic action explanatorily prior to the existence, say, of the executive

Free Creatures of the Universal Cause

63

state of intention the direct causing of which constitutes (or just is) that very action. For, whether we say that the agent’s basic action consists in just the causal relation between the agent and the intention, or whether we say that it consists in the intention together with the causal relation, the agent’s action will presuppose the intention. If the agent’s action consists in just the relation,31 then the action will presuppose the intention as a relation presupposes its relata. If the action consists in the relation and the intention together, then the action will presuppose the intention as that in which it partially consists. Either way, the agent’s basic action will not be explanatorily prior to the intention, since the action will presuppose the intention’s existence.32 In this way, God’s agency as characterized by EM is not in all respects sui generis. While both God and creaturely agent-causes cause, and therefore, explain, or account for, the effects they bring about in their basic causal acts, these basic acts do not themselves cause, nor are they explanatorily prior to, the effects brought about in those acts. With the foregoing in place, now is an opportune time to take stock of the multifaceted relationships that hold between God, A, and God’s act of causing A, assuming EM and DUC. Continue to let A stand for any creaturely act caused by God. As should now be clear, on EM, contrary to what is sometimes supposed, God, but not God’s act of causing A, is a cause of A. That God’s act does not cause A follows, again, from the fact that a cause must be prior to its effect, but God’s act is not prior to A. In causing or bringing about A, God simultaneously brings about his act of causing A, since that act just is his causing A. God is not logically sufficient for A, since there are worlds in which God exists and A does not. God is, however, logically necessary for A. For, A cannot exist unless caused by God, and A cannot be caused by God in any world where (per impossibile) God does not exist. God’s act of causing A is both logically necessary and logically sufficient for A. Conversely, A is both logically necessary and logically sufficient for God’s act. God’s act is logically necessary for A, since, given DUC, there is no world in which A exists without being caused by God. From the fact that God’s act is logically necessary for A, it follows that A is logically sufficient for God’s act. That God’s act is logically sufficient for A, and A logically necessary for God’s act, is clear for two reasons. First, as a matter of conceptual analysis, God causes A only if A gets caused, and A gets caused only if A exists. Second, God’s act at least partially and essentially consists in A. For both reasons, then, there is no world containing God’s act but not A. Given EM and DUC, it is not possible for God’s act of causing A to exist without A nor for A to exist without God’s act of causing A. Thus, the co-operation of God and the creaturely agent whose act A is, is necessary in order for either God or the creaturely agent to operate. The co-operation between God and the creaturely agent is one on which neither God’s act nor the creature’s act can be prior to the other. God’s act of causing A cannot be prior to A for reasons already discussed. A cannot be prior to God’s act of causing A, since God must already be causing A in order for A to exist. Since both God’s act and the creature’s act are necessary for the existence of the other, yet neither act is prior to the other, we can say that God’s act and the creature’s act are simultaneous (or concurrent) necessary conditions for each other.33

64

Free Will and God’s Universal Causality: The Dual Sources Account

The idea of two entities being simultaneous necessary conditions for each other might be thought problematic at first. Some might suppose that if a is a necessary condition for the existence of b, then a must be prior to, and not just simultaneous with, b. Others might worry that if a and b are necessary for the existence of each other, then since the existence of each would presuppose the existence of the other, the conditions for neither’s existence could ever be satisfied. These concerns, however, do not appear fatal to the idea of two entities serving as simultaneous necessary conditions for each other. With respect to the first concern, a need not be prior to b in order to serve as a necessary condition for b’s existence. All that is required is that b not exist in any world before or until a exists. But that b not exist in any world before or until a exists is consistent with a’s existing simultaneously with b, not just prior to b. Thus, the fact that neither A nor God’s act of causing A is prior to the other does not rule out their serving as necessary conditions for each other’s existence. With respect to the second concern, all that is precluded by a and b’s presupposing the existence of each other is that either be the cause of the other. For, since a cause brings it about that its effect exists in the first place, it can’t presuppose the existence of its effect at or prior to causing that effect. But there is no impossibility in a’s or b’s existing or coming to exist even though neither can exist without the other. All that is required is their existing or coming to exist simultaneously, which is precisely what happens in the case of A and God’s act of causing A. The foregoing should be enough to ward off the concern that if God’s act of causing A “presupposes” A, then A must already exist in order for God to cause it. Clearly, an effect can’t preexist its being caused. But on EM, God’s effects don’t preexist God’s causing them. We must distinguish two claims: 1. God’s act of causing A is not prior to A, and 2. A is prior to God’s act of causing A. (2) implies the absurd claim that A must already exist in order for God to cause it. But (1) does not. EM affirms (1), but rejects (2). On EM, God’s act of causing A is not prior to A. But neither is A prior to God’s act of causing it. Thus, A need not already exist in order for God to cause it. Rather, A and God’s act of causing it come to be simultaneously when God causes A. Finally, I have spoken of God’s act as essentially consisting in A together with A’s relation of dependence on God, and some might wonder whether that makes A prior to God’s act, contrary to what has been said above. I certainly do not intend “consisting in” to imply that A is prior to God’s act; indeed, it can’t be prior, since A can’t exist prior to God’s causing it. The locution “consists in” signals that God’s act of causing A just is A qua dependent on God. It does not mean that A and the causal relation are metaphysically prior components from which God’s act derives. For comparison, when a reductive materialist says that a mental state, such as a belief, “consists in” a brain state or states, he is not saying that the brain state or states are metaphysically prior to the mental state, as that out of which the mental state is constructed or composed. He is rather saying that the mental state just is the brain state(s) and nothing more.34

Free Creatures of the Universal Cause

65

4.5 Ability to Do Otherwise As we saw in Section 1.2, it is typically not the absence of determinism per se that matters most to libertarians, but rather certain other characteristics they take to be essential to free action and that they believe comprehensive determinism rules out. Two characteristics surface most often: i. An act’s being such that its agent had the ability to do otherwise ii. An act’s being ultimately up to its agent, something for which its agent is ultimately responsible. Since libertarians typically reject comprehensive determinism in order to safeguard such characteristics, it is important to show not just that DUC is compatible with creaturely acts being undetermined but also that it is compatible with these characteristics. In this and the following section, I attempt to do just that, arguing that on EM God’s causing our acts rules out neither our “ability to do otherwise” nor our “ultimate responsibility” as these characteristics have been understood by representative libertarians.35 We can begin by noting that not only libertarians and incompatibilists but also many compatibilists have maintained that free action requires the ability to do otherwise. Yet, the compatibilist understanding of this ability is consistent with the act in question being determined. In order to distinguish the libertarian from the compatibilist understanding, it is necessary to emphasize that on the libertarian but not the compatibilist view, ability to do otherwise means that the agent could have done otherwise all antecedent conditions remaining the same.36 On this understanding, an agent S who performs action A has the ability to do otherwise only if there is a possible world the same as the actual world in all factors or conditions prior to A, but in which S does other than A.37 That EM renders God’s causing our actions consistent with the libertarian understanding of ability to do otherwise can be seen as follows. Leave God out of the picture for a moment and suppose that we have an act such that its agent could have done otherwise all antecedent conditions remaining the same. In the actual world, W, agent S performs action A, and there is a possible world, W*, the same as W in all factors prior to A, but in which S does other than A. Now add that in W, A is caused by God. Will it cease to be the case that there is a possible world like W* the same in all factors prior to A, but where S does other than A? The answer, of course, is that there will cease to be a possible world like W* only if God’s causing A introduces some factor that is both prior to and logically sufficient for A. But, as we have already seen in the previous section, given EM, God’s causing A does not introduce such a factor. Consequently, God’s causing A is consistent with S’s having the ability to do otherwise in the libertarian sense, that is, all antecedent conditions remaining the same. It goes without saying, perhaps, that the mere fact that God’s act of causing A is (assuming DUC) a necessary condition, without which S does not perform A, does not rule out S’s having the ability to do otherwise in the libertarian sense; nor does the mere fact that God’s act constitutes something logically sufficient for A. Libertarians

66

Free Will and God’s Universal Causality: The Dual Sources Account

generally acknowledge an agent’s ability to have done otherwise than he did, even while acknowledging also that the agent’s doing what he did required that certain necessary conditions for the act were fulfilled—the continuing existence of the universe, for instance. And though my knowing my daughter is reading her book is logically sufficient for her so reading, libertarians will not typically conclude that my daughter could not have done otherwise than so read. Still, it might be argued that the libertarian should insist on the ability to do otherwise, not merely all antecedent conditions remaining the same but simply all conditions remaining the same. An agent S who performs action A has the ability to do otherwise all conditions remaining the same only if there is a possible world having in common with the actual world all conditions necessary for S’s performance of A, but in which S does other than A.38 While, given EM and DUC, God’s causing A is consistent with S’s having the ability to do otherwise all antecedent conditions remaining the same, it is not consistent with S’s having the ability to do otherwise all conditions remaining the same. For, since, given DUC, S does not perform A unless God causes A, a necessary condition of S’s performing A is God’s act of causing A. Add that God’s act of causing A is logically sufficient for A and does not occur in any world in which S does otherwise, and it follows that there is no possible world in which S does otherwise sharing with the actual world all conditions necessary for S’s performance of A. Thus, if libertarian freedom requires the ability to do otherwise all conditions, as opposed to antecedent conditions, remaining the same, then S won’t have the ability to do otherwise in the libertarian sense, after all. Yet, it would be a mistake to think that free action, libertarian or not, requires the ability to do otherwise all conditions, as opposed to antecedent conditions, remaining the same. For, given any agent S who performs act A, a factor or condition not prior to the performance of A exists too late to make S perform A, or to place a limit or restriction on S’s power with respect to A, such that S had to perform A and could not have done otherwise.39 To take our current example, God’s act is certainly a necessary condition for S’s performance of A. But, on EM, God’s act also presupposes A and therefore does not exist prior to S’s performing A. Since God’s act is not prior to, but rather simultaneous with, the exercise of S’s power, God’s act exists too late to make it such that S had to perform A and could not have done otherwise. By parity of reasoning, neither does S’s performance of A make it such that God could not have done otherwise than cause A. Since neither God nor S exercises his power prior to the other’s exercise, the abilities of both to do otherwise remain intact. Thus far, we have been considering a case where, in the actual world W, S performs A, asking whether DUC and EM are consistent with S’s having had the ability to refrain from so performing. Now suppose that, in the actual world W, S refrains. Are DUC and EM consistent with S’s having had the ability to perform A in W? It might seem not. For, given DUC, God’s causing A is a necessary condition of S’s performing A; there is no possible world in which S performs A without God’s causing it. Yet, S’s performing A is also a necessary condition of God’s causing it, such that God does not cause A in any world in which A is not performed by S. In the actual world W, S does not perform A, from which we can infer that God does not cause A. But, if God does not cause A, then a necessary condition for S’s performing A is absent in W. Won’t this mean that

Free Creatures of the Universal Cause

67

S lacked the ability to perform A in W? Indeed, given DUC and EM, won’t it be the case that in any world in which S does not perform A, S lacked the ability to perform A, since a necessary condition for S’s so performing—that God causes A—would have been absent? The correct answer, I contend, is “no.” Just as S’s ability to do otherwise than perform A is compromised only by factors prior to S’s performance of A that determine S to perform A or that prevent S from refraining, so S’s ability to have performed A, assuming S refrains, is compromised only by factors antecedent to S’s nonperformance that determine him not to perform A or that prevent him from performing A. A factor not prior to S’s refraining exists too late to make it such that S had to refrain or to place a limit or restriction on S’s power such that S could not have done otherwise than refrain. But, as we will see in the following paragraph, God’s not causing A is not prior to S’s not performing A. Thus, God’s not causing A does not take away S’s ability to have performed A in a case where S refrains. How can we judge whether one nonoccurrence, such as God’s not causing A, is prior to another nonoccurrence, such as S’s not performing A? At least the following principle seems right: The nonoccurrence (or nonexistence) of x is not explanatorily or causally prior to the nonoccurrence (or nonexistence) of y, if the occurrence (or existence) of x presupposes the occurrence (or existence) of y.40 For example, we cannot explain the nonexistence of Santa by the nonexistence of Santa’s risibility accident, since the existence of Santa’s accident presupposes the existence of Santa. Notice that, in this example, Santa’s risibility accident is a necessary condition of the existence of Santa. Any world without Santa’s risibility accident is a world without the jolly old elf.41 Nevertheless, the proposed explanans fails, because the nonexistence of one thing can’t explain the nonexistence of another, if the existence of the one presupposes the existence of the other. Since, then, God’s act of causing A presupposes S’s performing A, God’s not causing A cannot explain S’s not performing A. But, then, God’s not causing A is not explanatorily or causally prior to S’s not performing A. And so God’s not causing A exists too late to make it such that S could not have performed A in a case where S refrains. In sum, we have seen that whether S performs A or refrains from A, DUC and EM are consistent with S’s having had the ability to do otherwise all antecedent conditions remaining the same. This is the kind of “ability to do otherwise” that incompatibilists and libertarians have traditionally deemed necessary for free will. We have also seen that conditions or factors not prior to S’s performing, or refraining from, A exist too late to limit S’s power, making it such that S lacked the ability to do otherwise. Given DUC and EM, God’s causing A and S’s performing A are simultaneous necessary conditions for each other. There is no world in which S performs A without God’s causing A and no world in which God causes A without S’s performing it. What we have seen, however, is that from God’s not causing A, it doesn’t follow that S could not have performed A, only that S doesn’t. In the same way, from S’s not performing A, it doesn’t follow that God could not have caused A, only that God doesn’t. Given DUC and EM, both God and S retain the ability to do otherwise all antecedent conditions remaining the same. These conclusions will have significant implications in Chapters 7 and 8.

68

Free Will and God’s Universal Causality: The Dual Sources Account

4.6 Ultimate Responsibility Even if EM renders God’s causing our acts consistent with our ability to do otherwise, we need still to ask whether it renders consistent God’s causing our acts with those acts being ultimately up to us, such that we are ultimately responsible for them. Perhaps the most prominent detailed account of what is required for ultimate responsibility is Robert Kane’s UR condition. In the following three paragraphs, I will show that, given EM, an agent whose act is caused by God can satisfy Kane’s conditions for ultimate responsibility. Then, four paragraphs from now, I will give an independent and perhaps more straightforward reason for thinking that, assuming EM, an agent can be ultimately responsible for its actions, even if they are caused by God. According to Kane, “a willed action is ‘up to the agent’ in the sense required by free will only if the agent is ultimately responsible for it in the following sense”:42 (UR) An agent is ultimately responsible for some (event or state) E’s occurring only if (R) the agent is personally responsible for E’s occurring in a sense which entails that something the agent voluntarily (or willingly) did or omitted, and for which the agent could have voluntarily done otherwise, either was, or causally contributed to, E’s occurrence and made a difference to whether or not E occurred; and (U) for every X and Y (where X and Y represent occurrences of events and/or states) if the agent is personally responsible for X, and if Y is an arche (or sufficient ground or cause or explanation) for X, then the agent must also be personally responsible for Y.43

UR gives necessary and not sufficient conditions for an agent’s being ultimately responsible for something.44 That on EM an agent whose act is caused by God can satisfy these conditions can be seen as follows. Condition R is satisfied, because the agent’s act is something he voluntarily does and, as we have just seen, something for which he could have voluntarily done otherwise.45 Given that R is satisfied, U will fail to be satisfied only if there is some arche for the agent’s action for which the agent is not personally responsible in the sense indicated by R. What does Kane mean by an arche? Two characteristics appear essential for something to count as an arche as Kane understands it. First, an arche is a sufficient reason of one of the following three sorts: (i) a sufficient condition, (ii) a sufficient cause, or (iii) a sufficient motive. Kane tells us that archai of sorts (i) and (ii) entail, and are logically sufficient for, the existence of that for which they are sufficient reasons. Archai of type (iii), sufficient motives, do not strictly entail that for which they are sufficient reasons. But, having a sufficient motive for some act at a time means that, given the motive, performing the act at that time would be voluntary and omitting the act not voluntary.46 The second characteristic of archai is that they are origins, sources, or causes of that for which they are sufficient reasons.47 What precludes ultimate responsibility according to condition U of UR is that there be sources sufficient for our actions, for which sources we are not personally responsible in the sense indicated by R: “If the action did have such a sufficient reason for which the agent was not responsible, then the action, or the agent’s will to perform it, would have its source in something that the agent played no role in producing.”48

Free Creatures of the Universal Cause

69

With the meaning of arche made clear, we can now see that condition U is satisfied given EM. Given that condition R is satisfied, God’s causing an agent’s action will preclude U’s being satisfied only if it introduces an arche for the action for which the agent is not personally responsible in the sense indicated by R. We have already seen that on EM God’s causing some act does not introduce an origin, source, or cause of the act the obtaining of which entails or is logically sufficient for the act. Thus, God’s causing the act does not introduce any arche of sorts (i) or (ii) and hence does not introduce any arche of sorts (i) or (ii) for which the agent is not personally responsible. Nor if the agent’s act has an arche of sort (iii) does God’s causing the act preclude the agent from being personally responsible for that arche. For suppose that the act has a sufficient motive. God’s causing the act is perfectly consistent with its being the case that something the agent voluntarily did or omitted, and for which the agent could have done otherwise, causally contributed to the agent’s having that motive, and made a difference as to whether the agent had it. For instance, decisions that the agent made in the past, and for which he could have done otherwise, may have causally contributed to the agent’s having the sufficient motive he now has. It seems, then, that given EM, an act caused by God can satisfy Kane’s famous conditions for ultimate responsibility. Yet, I want to suggest further that we have independent reason to think that, on EM, an act caused by God can be ultimately up to its creaturely agent. For, it is sufficient for an act’s being ultimately up to its agent, I suggest, that it be within the agent’s power or control whether or not the act occurs. And it is sufficient for its being within the agent’s power or control whether or not the act occurs that the act be performed voluntarily and intentionally by the agent, with the agent’s having the ability voluntarily and intentionally to do otherwise all antecedent conditions remaining the same. On EM, as we have seen, God’s causing a creature’s act does not preclude the creature’s voluntarily and intentionally performing the act, having had the ability voluntarily and intentionally to do otherwise, all antecedent conditions remaining the same. Since, given EM, God’s causing a creaturely act does not remove that which suffices for its being within the creature’s power and control whether or not the act occurs, neither should we think it removes the act’s being ultimately up to its creaturely agent.49 Some may worry that if God causes my act, whether or not my act occurs will be ultimately up to God, and not me. This consequence might follow were it the case that both God’s causing my act is logically sufficient for my act’s occurring, and I have no say regarding whether or not God’s causing my act occurs. Yet, on EM, while God’s causing my act is certainly logically sufficient for my act, it is not the case that I have no say regarding the occurrence of God’s causing my act. On EM, God’s act of causing my act presupposes my act as that in which it partially and essentially consists. It follows that without my concurrent co-operation in performing my act, God’s act of causing my act does not occur. What’s more, none of the conditions on which my act depends, not even God’s causing my act, makes it such that I had to perform my act and could not do otherwise. Because God causes my act only with my co-operating when I could have done otherwise, I have a say as to whether God’s causing my act occurs. On EM, I exercise counterfactual power over God’s causing my act: Agent S has counterfactual power over event E if S performs some act with respect to which S could have done otherwise all antecedent conditions remaining the same, and without which act event E would not have occurred.

70

Free Will and God’s Universal Causality: The Dual Sources Account

Given EM, I exercise counterfactual power over God’s causing my act, since, all antecedent conditions remaining the same, I have the ability to do otherwise, and were I to do otherwise, God’s act of causing my act would not occur. Since whether God’s causing my act occurs is not outside my power, there is no reason on EM to think that whether or not my act occurs is ultimately up to God and not, at least also, ultimately up to me.

4.7 Dual Sources The Dual Sources approach to divine and creaturely agency consists in the coupling of DUC and EM together with the foregoing account of how EM enables us to reconcile DUC with libertarian creaturely freedom. According to Dual Sources, all our acts can be caused by God and yet all the standard conditions for libertarian freedom satisfied. God can cause our acts without determining them, thus making it possible, on both the strict and broad accounts, for our acts to be free in the libertarian sense. God can cause our acts without removing our ability to do otherwise all antecedent conditions remaining the same. And God can cause our acts and yet our acts be ultimately up to us, acts for which we are ultimately responsible. Yet, Dual Sources also maintains that our acts are ultimately up to God—that they are ultimately up to us and God—that between God and his free creatures, ultimate responsibility is not a zero-sum game. After all, God causes our acts, and it is not possible that they exist without God’s causing them. Moreover, for any effect God brings about, God could have done otherwise all antecedent conditions remaining the same. As a consequence, God, just like us, has control over whether or not our acts occur. And just as we enjoy counterfactual power over God’s causing our acts, so God enjoys counterfactual, as well as causal, power over our acts. On Dual Sources, our free acts are ultimately up to God and us. And that, I contend, is just what the libertarian proponent of DUC should wish to claim. Indeed, that our acts can be free in the libertarian sense and yet also ultimately up to God will be critical to showing, in Chapter 8, how Dual Sources enables robust accounts of providence, grace, and predestination, all consistent with libertarian creaturely freedom. Some might object that contrary to Dual Sources, libertarian freedom requires that our free acts be ultimately up to us and us alone—that they have no cause or source other than us. As a mere assertion, the objection begs the question against Dual Sources. The only principled support for the objection would be showing the truth of the following conditional: If all our acts are ultimately up to, or have a cause or source in, any agent or thing other than us, then none of our acts satisfy one or more of the standard conditions for libertarian freedom: that is, none is undetermined or free in the libertarian sense; and/or is such that we could have done otherwise all antecedent conditions remaining the same; and/or is such that we are ultimately responsible for it.

Free Creatures of the Universal Cause

71

Yet, we have seen that this conditional is false on standard understandings of the relevant terms. Admittedly, one can stipulate whatever definition of “libertarian” one wants. But I know of no reason to think that genuine freedom requires satisfying a revised definition of libertarianism on which our acts being caused by and ultimately up to God (as well as us) is ruled out, simply as a matter of definition. On the contrary, satisfying the standard conditions for libertarian freedom gives us all the freedom we need to make us morally responsible agents in control of our actions. Dual Sources enables us to satisfy these standard conditions, even while our acts are also caused by and ultimately up to God. To be sure, in order to have libertarian freedom, God has to create us with the power to act freely, including to act, at least sometimes, with the ability to do otherwise all antecedent conditions remaining the same. It was within God’s power not to have given us such freedom. Moreover, God could have made it the case that all our acts are determined by always causing some factor both prior to and logically sufficient for our acts. What Dual Sources shows is that if God chooses to give us the power of libertarian freedom, adding that God causes our actions does nothing to remove that freedom. Given EM, God’s causing our actions does not as such introduce any factor that determines them. We have seen that EM, so central to Dual Sources, is rooted in scholastic theology. More generally, Dual Sources has its inspiration in, and attempts to develop, a scholastic—primarily Thomistic—tradition of thinking about divine and creaturely agency. Proponents of this tradition emphasize the transcendence of the universal cause, the radical difference between God and creaturely agency, which precludes understanding divine action as a force that coerces, determines, or necessitates human action. Denys Turner spiritedly captures the thrust of this tradition when he contends: To say that my actions are free only insofar as God does not cause them presupposes a plainly idolatrous conception of the divine causality. … God’s action in the world is not a sort of cosmic elbow-jogging, so that where space is occupied by the divine action freedom is evacuated from it, and vice versa. That is a theology … [on which] God’s causality is construed as if it were just an infinitely scaled-up creaturely agency.50

Or consider Brian Davies’s characterization of Aquinas’s position, a position presumably shared by Davies himself: (1) Some things or processes in the world come about of necessity; (2) some do not; (3) yet both come about because of God’s creative activity, which is not to be thought of as like that of a creaturely cause that renders its effect inevitable (or determined or necessitated). If I place olive oil in a pan and put heat under it, and if nothing interferes with what I am doing, the oil will inevitably become hot. It has no choice. But, Aquinas thinks, there are in the world things with choice, albeit things created by God. These things are what they are and do what they do because God makes them to be as they are and to do what they do. But this making to be, unlike oil becoming hot because of a flame under it, does not coerce as a created agent acting on another might be thought to coerce.51

72

Free Will and God’s Universal Causality: The Dual Sources Account

Consider now Aquinas’s own diagnosis of the misstep in the inference from God’s universal causality to the conclusion that all is necessitated: But if God’s providence is the cause of everything that happens in the world … it seems that everything must happen the way it does: firstly, because he knows it and his knowledge cannot be mistaken, so what he knows must necessarily happen; and secondly, because he wills it and his will can’t be ineffective, so everything he wills, it seems, must necessarily happen. But these objections depend on thinking of knowledge in God’s mind and the working of God’s will on the model of such activities in us, when they are in fact very different.52

As he proceeds to spell out the difference between God’s will and our own, Aquinas states: God’s will is to be thought of as existing outside the realm of existents, as a cause from which pours forth everything that exists in all its variant forms. Now what can be and what must be are variants of being, so that it is from God’s will itself that things derive whether they must be or may or may not be and the distinction of the two according to the nature of their immediate causes. For he prepares causes that must cause for those effects that he wills must be, and causes that might cause but might fail to cause for those effects that he wills might or might not be. And it is because of the nature of their causes that some effects are said to be effects that must be and others effects that need not be, although all depend on God’s will as primary cause, a cause which transcends this distinction between must and might not. But the same cannot be said of human will or of any other cause, since every other cause exists within the realm of must and might not. So of every other cause it must be said either that it can fail to cause, or that its effect must be and cannot not be; God’s will, however, cannot fail, and yet not all his effects must be, but some can be or not be.53

Dual Sources does not constitute so much an interpretation of this difficult passage as it does a diagnosis and solution of similar spirit. If like PM we model divine agency on human agency, adding only that God is all-powerful such that his willing to cause something can’t be ineffectual, then God’s causing a creaturely act A (or any creaturely effect, for that matter) undeniably necessitates it. For, as we have seen, on PM, God’s causing A involves a divine act of choosing or willing to cause A that is both prior to A (since it is that in virtue of which God causes A) and logically sufficient for A (since it is not possible for God to will to cause A yet A not eventuate). Yet, the doctrine of divine simplicity and the doctrine that God is not really related to creatures—two doctrines that perhaps more than any others are associated with a strong view of God’s transcendence—recommend a different model of divine agency, namely, EM. And, on EM, God’s causing A does not introduce a factor both prior to and logically sufficient for, a factor that determines or necessitates, A. Thus, it is ultimately God’s transcendence that underlies the conception of divine agency that enables DUC to be reconciled with libertarian creaturely freedom.54

Free Creatures of the Universal Cause

73

Given EM, God’s causing a creaturely effect E does not determine or necessitate it. If God wills an effect to be necessitated, he must prepare a creaturely cause that necessitates it, just as Aquinas says above. Yet, even though God’s causing his effects doesn’t necessitate them, there is a clear sense on which, as Aquinas says, God’s will cannot fail (cannot be ineffective). For, on EM, as we’ve seen, God’s act of causing E is also an act of willing E. But, as we’ve also seen, God’s act of causing E is logically sufficient for E. Accordingly, on EM, there is a divine willing of E that is incompatible with the absence of E.55 Dual Sources has much going for it, yet more is needed to defend the account and exhibit its most important implications. In the following chapter, I will begin by offering a more extensive defense of EM, showing even further that the model is not ad hoc and that it can withstand key objections. Chapters 6 and 7 address what might be the greatest concern for many readers, the worry that even if it can preserve libertarian freedom, Dual Sources makes God the cause of sin and makes it impossible to respond to the problem of moral evil. I conclude in Chapter 8 by showing how Dual Sources provides robust accounts of God’s foreknowledge, providence, grace, and predestination, all consistent with libertarian creaturely freedom.

5

The Extrinsic Model Defended

We saw in Chapter 4 that the extrinsic model of divine agency (EM) enables us to reconcile DUC with libertarian creaturely freedom. This chapter defends the model, showing that EM is not ad hoc and that it can withstand key objections. That EM is not ad hoc should already be evident from the previous chapter, which illustrated its connection to central tenets of scholastic theology. In this chapter, I consider more broadly the alternatives to EM. I show that no alternative model can accommodate, or accommodate as well, certain central scholastic doctrines. I show, further, that DUC, together with plausible assumptions embraced by many theists (whether scholastic or not), actually implies EM. Far from its being ad hoc, EM is thus something to which the proponent of DUC is likely already committed. To secure this last point, I address an objection that EM is, itself, incompatible with DUC. I conclude by responding to an objection that EM renders divine causality unintelligible.

5.1 The Extrinsic Model, Intrinsic Models, and Scholastic Theology No account of divine agency other than EM can accommodate, or accommodate as well, certain central tenets of scholastic theology. Recall from the previous chapter what’s on the scene when God brings about a creaturely effect, E, on EM. There is: ● ● ● ● ●

●

God A God’s reason for causing A The causal-dependence relation between God and A God’s causal act, or causing of A, which consists in A plus the causal relation between God and A God’s willing or choosing A, which is nothing else than God’s causing A for a reason when God could have done otherwise.

I contrasted EM to a popular model of God’s agency (PM), on which, in addition to God, E, God’s reason for causing E, and the causal-dependence relation, God’s causing E also involves: ●

God’s choice, decree, or intention to bring about E, which is intrinsic to God, is that in virtue of which God causes E and which would not exist were God not causing E.

76

Free Will and God’s Universal Causality: The Dual Sources Account

PM is a particular example of a broader class of models of God’s agency, what I call “intrinsic models” of divine agency. An intrinsic model of divine agency is any model on which: i. God’s act of causing E is itself intrinsic to God; and/or ii. God’s causing E involves some choice, decree, or intention of God’s to cause E, which choice (etc.) is intrinsic to God; and/or iii. God’s causing E involves some other intrinsic feature of God that would not exist were God not causing E.1 If a model of divine agency is not an intrinsic model, then it is an “extrinsic model.” EM is just my preferred version of an extrinsic model. In fact, while EM has a number of recent defenders,2 I know of no one who has defended an extrinsic model other than EM. Intrinsic models of divine agency, however popular they may be among contemporary philosophers of religion, would appear to be ruled out by central tenets of scholastic theology. For example, as we saw in Section 4.3, the doctrine that God is not really related to creatures denies that relations between God and creatures have foundations in God. But, on an intrinsic model, at least one of (i)–(iii) must hold. And each of (i)–(iii) would seem to imply a foundation in God for God’s causal relation to E. If God’s act of causing E were intrinsic to God, then that would seem to constitute a foundation in God for God’s causal relation to E. God’s choice (etc.) to cause E would also seem to constitute such a foundation, especially if we add that God’s choice (etc.) cannot be impeded or ineffectual.3 And, likewise, any other feature involved in God’s causing E that would not exist were God not causing E would seem to constitute an intrinsic foundation in God for God’s causal relation to E. Apparently, then, the scholastic teaching on God’s relations rules out intrinsic models. Even more apparent is the conflict between intrinsic models and the scholastic doctrines of divine simplicity and divine freedom. Recall from Section 4.3 that divine simplicity denies both that God is composed and that there are any entities intrinsic to God that are distinct from (i.e., not identical to) God. But, if God causes E, any intrinsic model will include at least one entity intrinsic to God of the sorts referenced in (i)–(iii). Such an entity would either have to be identical to God, belong to God essentially without being identical to him, or reside in God as a contingent accident. But divine simplicity rules out God’s having any intrinsic entities distinct from him, and thus rules out the entity being an accident in God or belonging to God essentially without being identical to him. And, if God’s act of causing E or choosing to cause E were identical to God—or if there were something else identical to God that would not exist were God not causing E—then contrary to the common scholastic understanding of God’s freedom, God could not have done otherwise than cause, or choose to cause, E.4 The foregoing argument that intrinsic models are inconsistent with the scholastic doctrines of divine simplicity and divine freedom, admittedly, depends on an unstated assumption, namely, the truth of object essentialism. Acts, of course, take objects. If I make a pot of coffee, coffee is the object made. If I know that Rome is in Italy, that

The Extrinsic Model Defended

77

Rome is in Italy is the object of my knowing. Object essentialism is the claim that acts have their objects essentially: It is not possible for an act with a specified object to exist without that same object. An implication of this claim is that act a and act b can be the same or identical act only if they have the same object(s). Object essentialism has a strong intuitive pull. Suppose Kristen chooses to purchase ice cream at the market. Had she chosen cookies, instead, would she have performed the same act of choosing as the one she actually performed? I suspect most will say “no.” But the sole reason for thinking the act would have been different is that it has a different object. To agree that the act would have been different, then, is to agree that diversity of object signals diversity of act, which is just what object essentialism amounts to. Nevertheless, by rejecting object essentialism, one might attempt a scholastic intrinsic model (SIM), an intrinsic model compatible with the tenets of scholastic theology. Indeed, a number of scholastic authors have done just that.5 Such authors take God’s acts to be intrinsic to God, but avoid conflict with divine simplicity by denying that they are intrinsic accidents, maintaining instead that they are simply identical to God. Of course, one who identified God’s acts with God and affirmed object essentialism would have to give up the scholastic view that God could have done otherwise than he does. For, assuming object essentialism, God’s act of causing E or choosing to cause E could not exist without this same object. Thus, if God were identical to his acts of causing and choosing, then God could not exist without causing and choosing E, and so could not do otherwise than cause or choose E. This problem can be at least mitigated by denying object essentialism. Strictly speaking, SIM’s approach of identifying God’s actions with God while denying object essentialism would not enable us to say that God could have done or acted otherwise. For if God is identical to all his actions or activity, it is not possible for God to exist without those actions or activity. But denying object essentialism would enable us to say that the same activity that God is might have had different objects, and thus that God (even though he could not have acted or done differently) might have had different doings in the sense of different effects. If this latter is enough for divine freedom, then the rejection of object essentialism enables a proponent of SIM to maintain an intrinsic model of divine agency alongside divine freedom and divine simplicity. Proponents of SIM hold that God is identical, say, to intrinsic acts of causing or choosing to cause E. Yet, since these same acts might not have had E as their object, whether or not God can be characterized as causing or choosing E does not depend on how God is intrinsically. Rather, it depends on how God is related to items extrinsic to him, in particular, whether God stands in a causal relation to E. It is, of course, up to God whether he causes E and thus stands in this causal relation. The point is that, on SIM as on EM, what God is intrinsically does not determine whether God stands in a causal relation to E. Interestingly, like EM, SIM would enable a reconciliation of DUC with creaturely libertarian freedom. On SIM, God’s activity of causing or choosing to cause some creaturely act A causes A, since this activity is identical to God, who causes A. This activity, then, is causally prior to A. But, on SIM, this activity is not logically sufficient for A. That’s because, given the denial of object essentialism, this same activity might

78

Free Will and God’s Universal Causality: The Dual Sources Account

not have had A as its object. Thus, on SIM, God’s causing A, though prior to A, does not involve any factor both prior to and logically sufficient for A, and, thus, does not involve any factor that renders A determined.6 Moreover, on SIM, even though God’s activity of causing and choosing to cause A is prior to A, whether or not it has A as an object depends on whether God stands in a causal relation to A. Since relations are not prior to their relata, God’s causal relation to A is not prior to A. But if God’s causal relation to A is not prior to A, then whether or not God’s activity has A as an object is not prior to the creature’s performing A. And, if that’s true, then whether God’s activity has A as an object is not outside the creature’s control for reasons similar to those discussed in connection with EM in Section 4.6. Despite its attractions, SIM has some definite drawbacks. First, it is a liability of SIM that it cannot affirm literally that God could have done or acted otherwise—indeed, it would be, strictly speaking, false on SIM to say that God does or acts differently even in a world where he doesn’t create anything at all! By contrast, since (unlike SIM) EM does not identify with God God’s acts that take creatures as objects, it can affirm literally that God’s activity itself, and not just the effects of that activity, might have been different. Second, SIM’s denial of object essentialism also seems a liability. Just as it would be counterintuitive to say that Kristen’s choosing to purchase cookies could have been the same act as her choosing to purchase ice cream, so it is counterintuitive to say that God’s act of willing to create the universe could be identical to any act that is not an act of willing to create this universe—better to endorse a model, like EM, that has the advantages of SIM without the counterintuitive rejection of object essentialism.7 Finally, both SIM and EM agree that all the differences between the actual world in which God wills and creates the universe and a possible world in which he doesn’t are to be found in items extrinsic to God. That’s because in order to preserve God’s freedom, both models maintain that there are possible worlds in which the identical divine substance brings about different effects, while in order to preserve divine simplicity, both models deny that God has any intrinsic features distinct from the divine substance. Yet, it is surely odd to hold that all the differences between a world in which God wills and creates the universe and a world in which he doesn’t are extrinsic to God, yet claim that God’s willing and creating the universe are wholly intrinsic—indeed identical—to him. A proponent of SIM is committed to this incongruity. EM avoids it by acknowledging that God’s willing and creating the universe are extrinsic. But are there extrinsic models other than EM that are compatible with scholastic theology? As indicated above, I know of no one who has defended an extrinsic model other than EM. But the only other extrinsic models I can imagine anyone proposing are ones that also conflict with scholastic theology. For, as we will see in the following section, they conflict either with DUC, which as we know from Chapter 1 is itself a central tenet of scholastic theology, or with other assumptions embraced by scholastic theology. In short, the only extrinsic models I can imagine other than EM prove inconsistent with scholastic theology. Every intrinsic model other than SIM proves inconsistent with scholastic theology. And SIM, while consistent with scholastic theology, imports costs that are avoided by EM. Those committed to the central tenets of scholastic theology will need to accept either EM or SIM, and of these two, EM is preferable.

The Extrinsic Model Defended

79

The foregoing, of course, depends on seeing that EM itself can accommodate the basic tenets of scholastic theology. On EM, God’s acts of causing, willing, and choosing E are not intrinsic to God; nor do they involve any intrinsic feature of God that would not exist were God not performing these acts. For this reason, there is no violation of divine simplicity by attributing something intrinsic to God distinct from the divine substance. Nor, as we have seen, is anything identified with God that would preclude God’s ability to do otherwise. Nor again is there a violation of the scholastic teaching on God’s relations by attributing to God some intrinsic feature that would constitute a foundation in God for God’s causal relation to E. Nor, as we will see, does EM conflict with DUC in the way that many of the other models do. But do we find EM actually articulated by classical representatives of the scholastic tradition? While I don’t know any place in the tradition where EM is set out as explicitly as I have here, key elements of the model can certainly be found in passages from Aquinas. Most significantly, in Book II, Chapter 11, of the Summa contra gentiles, Aquinas asks whether something is said of God in relation to creatures. He answers in the affirmative, including among his examples a range of divine acts that take creatures as objects. Thus, God is spoken of in relation to creatures because He knows them, because they proceed and have their being from Him, and because they are moved by Him.8 Having established that to predicate such actions of God is to predicate of God relations to creatures, Aquinas then insists at the start of the immediately following chapter that these relations do not really exist in God. The passage is worth quoting at length: Such relations that refer to God’s effects cannot exist in Him really. For they cannot exist in Him as accidents in a subject, since there is no accident in Him, as was shown in Book I. Neither can they be God’s very substance. For, since, as Aristotle says in the Categories, relations are those which according to their very being have a certain reference to another, God’s substance itself would have to be referred to another. But that which is itself referred to another depends upon that other in a certain way, since it can neither be nor be understood without it. Thus, it would follow that God’s substance would depend on something extrinsic to it. And so God would not be, of Himself, the necessary being as was shown in Book I. Therefore, such relations do not really exist in God.9

There are two points to note here. First, when Aquinas denies that these relations really exist in God, the relations in question include divine actions of the sort sampled in the previous chapter (SCG 2.11), which take creatures as their objects, for he seems clearly to think that to predicate such an action of God is to predicate a relation of God to creatures. Thus, what he says here implies that divine acts such as causing, willing, or choosing E are not in God, just as EM claims they are not. Second, Aquinas’s judgment—that if acts taking creatures as objects were identical to the divine substance, then the divine substance could not exist without these creatures—appears to presuppose object essentialism and, thus, to suggest that Aquinas is taking object essentialism for granted. For, as we saw above, if object essentialism is denied, then

80

Free Will and God’s Universal Causality: The Dual Sources Account

the divine substance could be identical to acts that take creatures as objects without it following that the divine substance could not exist without the creaturely objects it actually (though not essentially) has. While some may suppose that action is an intrinsic state or event of an agent,10 Aquinas accepts the Aristotelian principle that the action of the agent is in the patient. In his analysis of motion in both the Physics and the Metaphysics, Aristotle locates the act of the mover in the motion of the thing moved.11 Aquinas, and scholastics generally, endorse this view,12 which Freddoso helpfully explains as follows: Suppose that all the prerequisites for an agent’s acting are satisfied in a given case. These include the agent’s having a sufficient power to produce a given effect in a properly disposed patient, the agent’s being appropriately situated with respect to the patient, the patient’s being properly disposed to receive the formal determination that the agent is ready to communicate, the absence of impediments, etc. Then what is the difference between the agent’s acting in such a case and its not acting? The common scholastic response is that the difference is just the coming to be of the relevant effect in the patient insofar as that effect is dependent on the agent. So no new entity need be added to the agent; instead, the action consists in something being added to the patient.13

Thus, says Freddoso, when the term “is acting” is predicated of the agent, we have a case of what the scholastics call extrinsic denomination, i.e., a predication which is such that its truth is grounded in the subject’s relation to a reality extrinsic to itself.14

Embodying this Aristotelian-Scholastic principle, EM locates God’s causal action precisely in God’s effects qua dependent on him, not unlike what Aquinas holds when he remarks that “creation is really nothing but a certain relation [of the creature] to God together with a beginning of existence.”15 On EM, God’s causal acts consist in God’s effects together with the causal-dependence relations between him and his effects, and not in anything intrinsic to God himself. And since God’s acts of willing and choosing his effects are not distinct from his acts of causing them, neither are these acts of willing and choosing intrinsic to God.16

5.2 From DUC to the Extrinsic Model We have seen that commitment to central tenets of scholastic theology gives one reason to embrace EM. I now show further that EM is implied by DUC together with a handful of assumptions, all of which are plausible, and many of which are widely shared by theists, whether scholastic or not. To show that DUC together with these assumptions implies EM, I need to show that DUC and these assumptions rule out all alternative accounts of divine agency.

The Extrinsic Model Defended

81

Implicit in the discussion of the previous section is that accounts of divine agency can be divided between intrinsic and extrinsic models, and between models that affirm object essentialism and models that deny it. For the most part, whether an account accepts or rejects object essentialism is not relevant to the question of whether it is ruled out by DUC and assumptions. Thus, in what follows, I will make explicit mention of object essentialism only in the one case where it might seem relevant. Let us begin with extrinsic models other than EM. As noted, I know of no one who has defended such a model. There are only two other extrinsic models I can realistically imagine someone defending, and, with respect to the second, it is difficult to imagine. The first of these alternative extrinsic models—call it AEM-1—is just like EM, with one exception. Unlike EM, AEM-1 holds that God’s reason for causing a creaturely effect E is logically sufficient for his causing E. But, as I will now show, no model of divine agency on which God’s reason for causing E is logically sufficient for his doing so is consistent with two assumptions shared not only by scholastics but by most contemporary theists. These assumptions are that God is essentially cognitively perfect, and that, in at least some cases, God could do otherwise than bring about what he does. To see the conflict, it will be helpful to define the notions of an “act context” and a “decision set.” For any act Φ that an agent S performs, the “act context” refers to the complete set of conditions antecedent to the performance of Φ, excluding the reasons known by S for or against performing Φ. The “decision set” refers to the reasons, known by S, for or against performing Φ in the given act context. Now, suppose in the actual world, God causes E for reason R. It will be the case either (a) that all possible worlds with the same act context are worlds in which God’s decision set includes R, or (b) that some worlds with the same act context are worlds in which God’s decision set does not include R. But (b) seems clearly incompatible with God’s essential cognitive perfection. Surely, it is not possible for an essentially cognitively perfect God to be oblivious to reasons for causing E, as God would be if the reason for which he caused E in the actual world were absent from his decision set in some possible world sharing the same act context. On the other hand, (a), when combined with the claim that God’s reason for causing E is logically sufficient for God’s causing E, conflicts with God’s freedom to do otherwise. For, if God’s reason for causing E is logically sufficient for his causing E, then there is no world in which God has that reason without causing E. And, given (a), there is no world in which God lacks that reason. Thus, contrary to a widely shared theistic belief, for any creaturely effect E that God brings about in a particular act context, God couldn’t have done otherwise than cause E in that same act context. The second alternative extrinsic model, AEM-2, is one on which, in addition to God, E, and the causal-dependence relation between God and E, God’s causing E also involves an entity extrinsic to God in virtue of which God causes E. Thus, we have: ● ● ● ●

God E The causal-dependence relation between God and E Some entity extrinsic to God in virtue of which God causes E.

82

Free Will and God’s Universal Causality: The Dual Sources Account

But no proponent of DUC can accept this model. For DUC holds that God directly causes all entities distinct from himself, which is to deny that God causes any such entity a by means of causing some other entity b which is a more proximate cause of a or which is that by means of which, or in virtue of which, God causes a. Yet, according to AEM-2, God can’t cause E except in virtue of some extrinsic entity. And since this entity is distinct from God, DUC requires that God also cause it. It follows, given both AEM-2 and DUC, that God can’t cause E except by means of causing the extrinsic entity in virtue of which he causes E. But, in that case, God doesn’t cause E directly in violation of DUC. Not knowing of alternative extrinsic models besides AEM-1 and AEM-2, let us turn to intrinsic models. Continuing to allow E to stand for any creaturely effect, recall that an intrinsic model of divine agency is any model on which either: i. God’s act of causing E is itself intrinsic to God; and/or ii. God’s causing E involves some choice, decree, or intention of God’s to cause E, which choice (etc.) is intrinsic to God; and/or iii. God’s causing E involves some other intrinsic feature of God that would not exist were God not causing E. The intrinsic item referenced in (i)–(iii), whether it be God’s act of bringing about E, or God’s choice (etc.) to bring about E, or something else, will either be identical to God, belong to God essentially without being identical to him, or belong to God as a contingent accident. Now, to identify the item with God or to say that it belongs to God essentially would clearly conflict with the widely held theistic assumption that for all or some of God’s creaturely effects, God was free to do otherwise than bring about or choose to bring about those effects.17 Thus, anyone committed to this assumption would need to reject versions of intrinsic models which make the relevant items identical to, or essential to, God. One might try to avoid the foregoing conclusion by denying object essentialism. But, as we saw in our discussion of SIM in the previous section, denying object essentialism while maintaining that the act of causing or choosing to cause E is identical to God still won’t enable one to affirm literally that God could have done or acted otherwise; only that the same activity that God is might have had different objects. Moreover, as we also saw, the denial of object essentialism is counterintuitive. It is counterintuitive to claim that an act of choosing E could be identical to any act that is not an act of choosing E. Those wishing to retain both object essentialism and the belief that for some of his creaturely effects God could literally have done otherwise will, consequently, need to reject intrinsic models that take the relevant items to be identical, or essential, to God. That leaves us to consider intrinsic models that take the relevant items to be intrinsic contingent accidents of God. Of these types of intrinsic models, we can further distinguish between models on which the relevant intrinsic accident is that in virtue of which God causes E, and so is explanatorily prior to E, and models on which the relevant intrinsic accident is not explanatorily prior to E, but merely accrues to God simultaneously with God’s causing E.

The Extrinsic Model Defended

83

What I have called the popular model is an example of the former type. PM holds that God’s choice (decree, etc.) to bring about E is that in virtue of which God causes E. But, however popular this sort of model may be, it cannot be combined with DUC for precisely the same reason that AEM-2 couldn’t. Given DUC, God must cause the intrinsic accidental entity in virtue of which he brings about E, since this entity is distinct from God. But that means God can’t bring about E except by means of bringing about the entity in virtue of which he brings about E. This type of model thus conflicts with DUC’s claim that God brings about his effects directly. We are left, then, to consider the last alternative to EM, namely, an intrinsic model that takes the relevant intrinsic item to be a contingent accident of God that is not explanatorily prior to E but that accrues to God simultaneously with God’s causing E. An example of this type of model would be one that holds God’s act of causing E to be an intrinsic accident of God, not prior to E, but acquired by God simultaneously with his standing in a causal relation to E. Another example of this type would be one on which God’s choice (etc.) to cause E accrues to God at the moment God causes E without being prior to E. I maintain that this type of model, together with plausible assumptions, also conflicts with DUC. In fact, coupled with these assumptions, DUC actually rules out God’s having intrinsic accidents at all and so rules out not only this type of model but also the previous type on which an intrinsic accident is that in virtue of which God causes E. Admittedly, the additional assumptions needed along with DUC to rule out God’s having intrinsic accidents are less obvious and so perhaps more controversial than those utilized in the previous arguments of this section. I believe the assumptions plausible. But let me point out that even if one thinks the following argument fails, intrinsic models on which an intrinsic accident accrues to God simultaneously (and not prior) to his causing A are, like EM and SIM, models on which DUC can be reconciled with libertarian creaturely freedom. For if such an accident—whether it be identified with God’s act of causing A, or God’s choice to cause A, or something else—is not prior to God’s standing in the causal relation to A, then neither is it prior to A, since relations are not prior to their relata. And if such an accident is not prior to A, then it does not constitute a factor both prior to and logically sufficient for A, and, hence, does not constitute a factor that renders the creature’s performance of A determined or necessitated. Moreover, since God’s coming to stand in a causal relation to A is not prior to the creature’s performance of A, for reasons similar to those discussed in Section 4.6, it would not be outside the creature’s control whether or not God has the intrinsic accident that accrues to God only when God comes to stand in that relation. Having made this point, let us turn to the argument for the incompatibility of DUC and God’s having intrinsic accidents. I noted when introducing the doctrine of divine simplicity in Section 4.3 that scholastic philosophy understands a thing’s intrinsic accidental properties to enter into its composition.18 This scholastic understanding seems correct in light of the fairly conventional way I defined the intrinsic–extrinsic distinction in Section 4.1. There I said that an entity is intrinsic with respect to an object if and only if the entity’s ceasing to exist or ceasing to belong to the object would of itself constitute either the object’s ceasing to exist (if the entity were identical to the object or belonged to the object

84

Free Will and God’s Universal Causality: The Dual Sources Account

essentially) or the object’s undergoing a real, as opposed to merely Cambridge, change. An entity not intrinsic with respect to an object is extrinsic with respect to that object. It follows from this definition that an accidental property is intrinsic with respect to an object only if the object’s losing the property would constitute its undergoing a real change. I take it that if a thing undergoes a real change, then, though it persists through the change, it is in itself really different after the change than it was before. If a thing differs only in its relations to things distinct from itself, without differing in itself, then the thing has not undergone real change, but merely Cambridge or relational change. Although such a characterization of real change does not seem especially controversial, when coupled with our equally common characterization of the intrinsic–extrinsic distinction, it yields substantive conclusions about how we should understand the relationship of a thing to its intrinsic accidental properties. In particular, it suggests that we should follow the scholastic tradition in understanding such properties to enter into the makeup or composition of a thing. To appreciate this implication, it is helpful to consider two different answers to the question, “What is included within the boundaries of an object to which intrinsic accidental properties belong?” Take Socrates, for instance. Does the object Socrates include only a certain subject or substance, excluding any accidental properties to which this substance may be related by way of inherence or exemplification? Or does the object Socrates encompass both Socrates’s substance and whatever accidental properties inhere in that substance at the time in question?19 If the former, then it is difficult to see how Socrates’s gaining or losing an accidental property constitutes a real change in Socrates. For if Socrates is only a substance, excluding any accidental properties to which that substance may be related, then gaining or losing an accidental property won’t of itself make Socrates any different than he was before he gained or lost the property. Certainly, Socrates will have acquired or lost a certain relation to a property. But gaining or losing a relation does not constitute a thing’s undergoing real change, unless a change in that relation of itself makes the thing different than it was before. And supposing Socrates encompasses just a substance, excluding any accidents, a change in Socrates’s relation to an accidental property won’t make Socrates himself any different. For, given this supposition, all the changes will fall outside the object that is Socrates. And, surely, Socrates doesn’t differ in light of changes all of which occur outside him. When generalized, the foregoing shows that no property not included within an object is a property the loss of which constitutes a real change in that object. But we also know, from our characterization of the intrinsic–extrinsic distinction, that no property the loss of which does not constitute a real change in an object is intrinsic with respect to that object. It follows that no property not included within an object is intrinsic to that object. If a property is intrinsic, it is thus included within the boundaries of the object to which it belongs. But an intrinsic property is not only included within its object’s boundaries. It is also a part of the object’s metaphysical makeup or composition; it is a component part of the object to which it belongs. By “component part,” I mean a part on which the object depends, and from which it partially derives, at the moment it has the part.

The Extrinsic Model Defended

85

Some objects may have some parts that are not “component parts” in this sense. But it seems to me that intrinsic properties count as component parts. For, it is at least partially in virtue of an object’s intrinsic properties that the being of the object takes the shape that it does at a particular time. And so the direction of dependence runs from the property to the object. Or, at the very least, it seems that an object at a time could not be prior to the intrinsic properties in virtue of which the being of the object takes the shape that it does at that time. The foregoing claim—that the direction of dependence runs from the property to the object—presupposes the conclusion argued for above, that an object with intrinsic properties is not simply a substance but includes both the substance and the intrinsic properties that belong to the object. The foregoing claim is, thus, consistent with the traditional scholastic view that intrinsic accidents depend for their existence on the substance in which they inhere. Intrinsic accidents do depend on the substance in which they inhere, but the object which depends on these accidents is not just the substance; it is, rather, an object composed of the substance and these accidents or at least composed of them at the time it has these accidents. With the foregoing considerations in place, we can finally see why DUC is incompatible with God’s having intrinsic accidents. Given what has been said, if God has intrinsic accidental properties, then he is composed, including among his component parts at least those properties and the divine substance. Now, the parts of an object are distinct from (not identical to) that object. And DUC states that God causes all entities distinct from himself at any time such entities exist. So, given DUC, God must cause his parts, if he has them. But a thing cannot simultaneously cause the parts that compose it at a time. For a cause is logically prior to its effect, at least at the moment it brings about its effect. But an object depends on the parts that compose it at a time and so cannot be logically prior to those parts at that time. Given that an object cannot simultaneously cause the parts that compose it, DUC requires us to deny that God has component parts.20 Since intrinsic accidental properties are component parts of the objects to which they belong, it follows that DUC rules out God’s having intrinsic accidental properties and thus rules out any model of divine agency that assumes such properties. Before concluding this section, let me address some possible concerns about the foregoing line of argument. I have argued that gaining or losing an accidental property does not constitute a real change in an object unless the object includes whatever accidental properties it has at a time. But it may seem that what I have argued also undermines the possibility of an object’s undergoing real change. For, an object cannot undergo real change unless it persists through the change. But two things I have said might seem to call into question the possibility of an object’s persisting. First, I have maintained that an object includes in its composition whatever accidental properties it has at a time. But, for example, if Socrates is currently sitting and the object Socrates includes a “sitting” accident, then it may appear as if anything without that “sitting” accident would not be Socrates and thus that Socrates cannot survive losing that accident. Second, I have maintained that objects with intrinsic accidents depend on those accidents at the time they have them. Yet, if so, it may seem that Socrates couldn’t continue if he lost his “sitting” accident, since that accident is something on which he depends.

86

Free Will and God’s Universal Causality: The Dual Sources Account

The first concern presupposes that an object, say, Socrates, is identical to whatever composition he has at a particular time, for instance, a composition that includes Socrates’s substance and a “sitting” accident. Only given this assumption would it follow from the composition’s ceasing to be that Socrates ceases to be. The concern is answered by denying that Socrates is identical to any particular composition. For Socrates to persist through change is not for any particular composition to persist, but rather for a certain rational animal to persist. Though the composition of this animal differs from moment to moment, it is the same animal in virtue of a single substance that underlies and is shared by these different compositions. Socrates’s substance is, therefore, that in virtue of which Socrates’s persists through change, while Socrates is really different after a change in virtue of no longer including within his makeup certain accidents that he had before. The second concern presupposes that an object’s depending on whatever parts enter its makeup at a given time implies that the object can’t exist without those parts. The solution is to deny this implication. It is true that Socrates depends on the parts that compose him at a given instant. He depends on them because it is in virtue of them that his existence has its present shape. It doesn’t follow, however, that he can exist only if he has these particular parts. As we’ve just seen, he can exist with different accidents in virtue of a persisting substance. Because it is commonality of substance that secures Socrates’s identity across various compositions, we might even use the term “substance” to categorize Socrates as the sort of thing that can persist through change, though we should be careful not to think of the object that Socrates is as merely a subject of accidents, excluding whatever accidents he has at a time.21 The key point in response to the second concern is that Socrates has some parts, some of his accidents, which he does not need in order to exist, even though he depends on the accidents he has at the time he has them. Because he does not need certain accidents in order to exist, he can persist even if he loses these accidents.22 One final concern: I have argued that an object’s intrinsic accidents are included among its component parts at the time it has those accidents, and also that an object cannot simultaneously cause whatever component parts it has at a time. It follows that an object cannot simultaneously cause one of its intrinsic accidents at the time it has it. But this implication may seem counterintuitive. For we might think that Socrates is causally responsible for his sitting and thus that he causes his sitting accident. This final concern admits of two solutions. The first solution is to say that the cause of the accident is, indeed, Socrates, but Socrates as he is composed a moment before acquiring the accident, not Socrates as he is composed during the time at which he has the accident. A proponent of this solution must be prepared to allow that a thing can, qua cause, precede its effect in time. A second solution is to say that, strictly speaking, it is not the whole of Socrates but some controlling part of Socrates, perhaps Socrates’s will, that causes the accident at the time Socrates has it. Since it is natural enough to hold Socrates causally responsible for what his will effects, this solution does justice to our intuition (if we have one) that Socrates is causally responsible for his sitting accident, even though, strictly speaking, it is Socrates’s will that is the cause. This second solution recalls Aristotle’s teaching that for an animal (or anything else) to move itself is for one part of the animal to effect

The Extrinsic Model Defended

87

motion in another part. A whole can thus bring about an effect in itself in virtue of one of its parts doing so.23 Thus concludes my defense of the claim that DUC, together with plausible assumptions, rules out any account of divine agency on which God has intrinsic accidents. At this point, all the alternatives to EM have been considered, and none is compatible with DUC and plausible assumptions. Far from its being ad hoc, the proponent of DUC has strong reasons to endorse EM.

5.3 But Is the Extrinsic Model Also Ruled Out by DUC? Of course, if EM were similarly ruled out by DUC, then it would undermine the force of the argument from DUC (and assumptions) to EM. Indeed, DUC can’t be true if there is no account of divine agency with which it is compatible. Thus, if EM is also ruled out by DUC, then we know either that DUC is false, or that we have not considered all the alternatives to EM, or that one of the other assumptions of the foregoing argument is false. It is important, therefore, to show that EM is not likewise ruled out by DUC. But it may seem that EM is ruled out. According to EM, God’s causing some creaturely effect E involves, among other things: ● ● ●

God E, and The causal-dependence relation between God and E.

DUC requires that God cause all entities distinct from himself. But the causaldependence relation between God and E appears to be an entity distinct from God. Thus, DUC requires that God cause this relation, call it R. To cause R, however, would seem to introduce a new causal relation R* that holds between God and R. And, since R* is distinct from God, DUC requires that God cause it, as well, introducing yet another causal relation R** that holds between God and R*, and so on to infinity. In short, it looks as if the coupling of DUC and EM leads to an infinite regress, according to which God must engage in an infinite number of causal acts in order to cause any creaturely effect E. Very plausibly, such a regress is vicious and untenable. Thus, it looks as if DUC rules out EM, too. Fortunately, the argument that the conjunction of DUC and EM leads to an infinite regress can be resisted, as we can see by considering three possible accounts of the ontology of relations. On one account, relations are entities that exist extrinsic to and between their relata, as a kind of medium connecting them. I will refer to this account as the medium view of relations. The medium view allows that a pair of relata, x and y, may be related to one another in virtue of features intrinsic to x and/or y. For example, if Elizabeth is taller than Cecilia, and Cecilia shorter than Elizabeth, these relations hold in virtue of Elizabeth’s and Cecilia’s heights. The medium view denies, however, that these relations are identical to either height taken separately or the heights taken together. Nor does the medium view think of these relations as distinct from their foundations (the heights), but nevertheless intrinsic to their relata (Cecilia

88

Free Will and God’s Universal Causality: The Dual Sources Account

and Elizabeth). For, a proponent of the view wants to allow that Elizabeth could cease being taller than Cecilia simply through a change in Cecilia’s height, without Elizabeth undergoing any real change herself. But if the relation were intrinsic to Elizabeth, then it would seem impossible for her to lose that relation without really changing. The medium view, thus, thinks of relations as entities distinct from anything intrinsic to their relata.24 On a second account of the ontology of relations, more common among scholastic authors, when two things are really related to one another, each relatum possesses two intrinsic entities: first, there is some non-relational entity that serves as the foundation for the relatum’s relation to the other, and, second, there is the actual relation to the other, which is ontologically distinct from the foundation. To take our example above, when Elizabeth is “taller than” Cecilia, this second account posits in Elizabeth not only her height as the foundation for the relation but also a distinct “taller than” relation, understood as an intrinsic item not identical to the foundation. Similarly, the account posits in Cecilia both her height, the foundation, and a distinct “shorter than” relation to Elizabeth. Because this approach understands a relation to be ontologically distinct from its foundation, and because it was popular among some scholastic authors, I will refer to this approach as the non-reductive scholastic view.25 On a third account of the ontology of relations, also espoused by scholastic authors, relations are not entities that exist between and extrinsic to their relata. Rather, a thing’s intrinsic features of themselves relate that thing to other things provided that certain conditions are satisfied. For example, Elizabeth’s height of itself relates her as “taller than” Cecilia under conditions in which Cecilia exists with a height that is less than Elizabeth’s. Another way of putting the view is to say that, under such conditions, Elizabeth’s height acquires the ratio (the intelligible character) of a “taller than” relation to Cecilia or that, under such conditions, Elizabeth’s height constitutes her relation of being taller than Cecilia. Under conditions in which Cecilia does not exist with a height that is less than Elizabeth’s, Elizabeth’s height does not, of course, relate her as “taller than” Cecilia, possess the ratio of such a relation, or constitute such a relation. Elizabeth can therefore lose the relation of being taller than Cecilia simply in virtue of a change in Cecilia’s height without Elizabeth undergoing any real change in herself. In such a circumstance, Elizabeth’s height remains the same, but ceases to relate her as taller than Cecilia, because the conditions under which her height constitutes that relation are no longer satisfied. For example, in conditions where Cecilia has grown taller than Elizabeth, Elizabeth’s height will relate her as shorter than, rather than taller than, Cecilia. Proponents of this view will speak of Elizabeth’s height as the “foundation” for the potential relations to other things of being taller than, shorter than, or the same height as these other things. Yet, when conditions are satisfied such that Elizabeth is actually related to something else in any of these ways, the relation is not an entity distinct from this foundation (Elizabeth’s height). Because, on this account, x’s relation to y is identical to the foundation in x under certain conditions (or to the intrinsic feature of x that relates x to y under certain conditions), I will refer to this account as the reductive scholastic view or the reductive view, for short.26 In introducing the reductive view, we have used Elizabeth’s height, an accidental property, as our example of a foundation that, under certain conditions, relates

The Extrinsic Model Defended

89

Elizabeth to, possesses the ratio of a relation to, or constitutes her relation to Cecilia. It is important to note, however, that on the reductive view anything intrinsic with respect to a subject, including the subject itself, can be a foundation for a relation and thus can actually relate the subject under the right conditions.27 We should also recall from Section 4.3 that when x is only rationally, and not really, related to y, there is no foundation in x for x’s relation to y. As a consequence, x’s rational relation to y is not anything intrinsic to x, as it is when x is really related to y with x’s relation to y being constituted by the foundation in x under the requisite conditions. In a case where x is rationally related to y, statements that predicate a relation of x to y are true simply in virtue of x, y, and whatever foundation in y constitutes, in the conditions, y’s relation to x.28 Given the reductive view of relations, it should be clear how we can resist the argument that the conjunction of DUC and EM leads to an infinite regress. For the reductive view enables us to deny that the cause–effect relation between God and E is an entity distinct from E. On the contrary, we can say that E relates itself to God as causally dependent on him or that E possesses the ratio of such dependence.29 Since the causal relation between God and E is not an entity distinct from E, God’s causal act on this view consists simply in E qua dependent on God, where the locution “qua dependent on God” connotes that E owes its existence to God, but does not correspond to any additional entity, E’s dependence relation, which is distinct from E. Because, on this view, there is no causal-dependence relation distinct from E, DUC does not require an additional divine causal act to cause this relation. Thus, the argument to an infinite regress never gets off the ground. What about E makes E of itself related to God as causally dependent on him or makes it such that E has the ratio of dependence on God? The best answer, I think, is to say that E’s mode of being is such that it must be caused to exist by something with a mode of being of a sort that God uniquely enjoys. Theists may differ in precisely how they characterize these diverse modes. Aquinas and Suarez do so by means of the distinction between being by participation and being per se, discussed in Section 2.5. There can only be one instance of being per se—the being we call “God.” Everything else that exists is a being by participation. And every being by participation is such that it must be immediately caused by the one being per se. It is because any creaturely effect is a being by participation that it has the ratio of dependence on the unique being per se and thus relates itself as causally dependent on God. The conditions for any creaturely effect’s actually so relating itself are always satisfied, since a being by participation is necessarily dependent on the unique being per se and so cannot exist without such dependence. It might be thought incoherent to say that a thing relates itself to another as causally dependent on that other. For, it may seem that a thing’s existing is logically prior to its being able to relate itself to anything. But, arguably, nothing causally dependent has an existence that is logically prior to that dependence. So, it looks impossible for a thing to relate itself as causally dependent on another and thus impossible for a creaturely effect E to relate itself to God as causally dependent on him. Fortunately for the reductive view, the foregoing argument is flawed in its claim that a thing’s existence is always logically prior to its being able to relate itself. The claim

90

Free Will and God’s Universal Causality: The Dual Sources Account

appears to assume that a thing’s relating itself to another, or having the ratio of being so related, is the same (or similar) to its performing a certain kind of action. But, while it is plausible to say that a thing is logically prior to any action it performs (at least, if that action is distinct from the thing), for a thing to relate itself to another is not for that thing to perform an action. A thing’s “relating itself to another” just means that, under certain conditions, the thing suffices for the truth of a statement predicating of that thing a relation to another, with the consequence that the truth of the statement does not require a relation distinct from the thing. And while the truth of such a statement certainly does require that the thing exist, it does not require that the thing exist, in any sense, prior to its being related. Thus, the foregoing argument fails to show any incoherence in a thing’s relating itself to another as causally dependent on that other and so fails to show any incoherence in E’s relating itself to God as causally dependent on him. The reductive view of relations enables us to resist the argument that the conjunction of EM and DUC leads to an infinite regress because it enables us to deny that God’s causing E introduces a relation distinct from E. If God’s causing E introduces no relation distinct from E, then there is no need for God to engage in an additional act of causing this relation, which would introduce yet another relation, requiring yet another divine causal act, and so on to infinity. Unlike the reductive view, the medium and non-reductive scholastic views of relations hold that God’s causing E does introduce a real relational entity distinct from E. In the case of the medium view, this entity is a causal relation holding between God and E, though intrinsic to neither. In the case of the non-reductive scholastic view, this entity is E’s real and distinct relation of causal dependence on God.30 On either view, since for God to cause E involves a causal or causal-dependence relation distinct from E, the infinite regress may appear unavoidable. I think the regress can be avoided, however, on either account. The key to avoiding the regress is affirming the proposition (i) that in directly causing or bringing about an effect, an agent-cause also causes or brings about the causal relation that holds between it and the effect. If an agent brings about the causal relation in the very act of bringing about an effect, then it brings about that relation without needing any additional causal act to do so. It follows that God can bring about the causal-dependence relation between God and E in the very act of causing E. And, if God can do that, then DUC will not require that he perform an additional act to cause that relation, and the regress never gets off the ground.31 If we assume that an agent-causal act consists in an agent’s causal relation R to an effect, then to imply, as does (i), that an agent does not need any additional causal act in order to bring about R is to imply that the agent’s bringing about R does not introduce another causal relation R* holding between the agent and R. In bringing about an effect, the agent brings about its causal relation to that effect (or the effect’s causaldependence relation on it), but it does so without introducing a new causal relation. The situation is, perhaps, analogous to what we might wish to say about other sorts of relations. Just as we might deny that for a thing x to “have” a relation R to something y introduces a new “having” relation R* between x and R, so by endorsing (i) we deny that an agent’s causing the causal relation R to its effect introduces a new causal relation R* between the agent and R.32

The Extrinsic Model Defended

91

(i) seems very plausible. To begin with, it seems right to say that in directly bringing about an effect an agent is, in that very same act, making the effect causally dependent on it and, thus, bringing about the causal-dependence relation between it and the effect. Furthermore, accepting (i) enables us to affirm two points regarding an agent and its action that seem quite natural to affirm: (ii) that a basic agent-causal act consists in an agent’s causal relation to an effect33; and (iii) that the agent is the cause of its act.34 If we deny (i), then we cannot affirm (ii) and (iii) together. For, if the agent’s causal act consists in the causal relation to the effect, the agent can’t cause his act unless he causes the relation. But, if we reject (i), then we can’t affirm that the agent brings about the causal relation without committing ourselves to an infinite regress. So, assuming that such a regress is untenable, we must endorse (i) in order to affirm both (ii) and (iii). As discussed in Section 3.5, some authors, such as O’Connor, deny that an agentcause brings about its act at least in part for fear that it leads to an infinite regress. The regress can be avoided, however, simply by allowing (i). For, in that case, given (ii), the agent will have brought about his act and will have done so without having to perform any additional causal act that would launch the regress.35 Even for those ready both to reject DUC and to deny that an agent brings about its causal act, there is an argument for accepting (i). This argument has five main assumptions: (iv) that it is untenable to hold that an agent’s bringing about an effect requires an infinite number of prior agent-causal acts; (v) that every effect that is brought about is brought about by an agent; (vi) that some agent-causal relations are contingent; (vii) that every contingent entity has a cause that brings it about; and (viii) that relations that take relata exist only if their relata exist.36 Taken together, these assumptions, (iv)–(viii), are incompatible with the denial of (i). For, given (vi) and (vii), there is an agent-causal relation R that is contingent and that has a cause that brings it about. Given (v), the cause of R is an agent. Given the rejection of (i), R is not brought about in the causal act of which it is a part by the agent of that act. Thus, R is brought about by an agent performing a distinct causal act from the act of which R is a part.37 This distinct act will include a causal relation R* that holds between the agent-cause of R and R. Given (viii), R* will exist only if R exists. Thus, given that R is contingent, R* is contingent. But, then, by (v) and (vii), R* is caused by an agent. And, given the rejection of (i), R* is brought about by an agent performing a distinct causal act from the act of which R* is a part. And this distinct causal act will include yet another causal relation R** that holds between the agent cause of R* and R*. And, since R* is contingent, so also R**, which, by (v) and (vii), will require R** to be caused by an agent, and so on to infinity. Such an infinite regress is contrary to (iv). Thus, (iv)–(viii) and the denial of (i) form an inconsistent hexad. Given that the conjunction of (iv)–(viii) is incompatible with the denial of (i), we must either accept (i) or reject one of (iv)–(viii). To my mind, the acceptance of (i) is considerably more appealing than rejecting any of (iv)–(viii). Others may agree. (iv), I assume, would be widely accepted. It seems, for example, behind O’Connor’s denial that agent-causes cause their own acts (see Section 3.5). Were an infinite number of prior acts necessary in order for an agent to perform a causal act, it is doubtful that the agent could cause anything at all. (v) will be endorsed by those who take the world’s real causal producers to be agents rather than events; certainly, those in the

92

Free Will and God’s Universal Causality: The Dual Sources Account

Aristotelian or scholastic tradition would endorse (v). (vi) cannot be denied by anyone who acknowledges that there are some instances of agent-causation that, though existing, might not have. (vii) has been endorsed by many, even if not all, philosophers. And (viii) is highly intuitive and would, I think, be widely endorsed. In short, those who find the acceptance of (i) more plausible than the rejection of any of (iv)–(viii) will have reason to affirm (i), even if they are prepared to reject DUC or the claim that agent-causes bring about their own actions. Given (i), the conjunction of DUC and EM need not lead to an infinite regress, even if we assume the medium or non-reductive scholastic view of relations. For God can bring about the causal-dependence relation between himself and E in the very bringing about of E without the need to perform an additional causal act. Where are we at the end of this section? We have learned that on at least three possible accounts of the ontology of relations, the medium view, the non-reductive scholastic view and the reductive scholastic view, the coupling of EM and DUC need not lead to an infinite regress. Given the arguments of the previous section, we can conclude, then, that DUC (together with plausible assumptions) rules out all contenders for an account of divine agency, except EM. Far from its being ad hoc, the model of divine agency that proves so useful in reconciling DUC with libertarian creaturely freedom is a model that the proponent of DUC should accept even apart from its usefulness. Before turning to consider another objection to EM in the final section of this chapter, let me say that, for the purposes of Dual Sources, EM can be construed according to any of our three accounts of relations. While I speak of a divine causal act’s consisting in God’s effect together with the cause–effect relation between God and God’s effect, I allow that the relation and the effect may be only conceptually distinct, as the reductive view would have it. On the reductive view, though we can conceptually distinguish God’s effect from its dependence on God, the dependence is not a distinct relational entity over and above the effect. On the medium view and the non-reductive scholastic view, by contrast, an effect’s causal-dependence is a distinct relational entity, the cause–effect relation that holds between the effect and its cause. Fortunately, EM can carry out its work in this book construed in terms of any of these three accounts.38

5.4 Does the Extrinsic Model Render Divine Causality Unintelligible? Nothing can exist without what belongs to it essentially. It follows that if a thing can exist differently than it does, it will be because the thing can have different intrinsic accidents or perhaps different matter. But (questions of the Incarnation aside) all theists agree that God is not material. And we saw in Section 4.3 that according to divine simplicity, God has no intrinsic accidents, a point I also argued for in Section 5.2 from DUC and plausible assumptions regarding change and composition. One who accepts divine simplicity and/or the argument from Section 5.2 must therefore deny that God has intrinsic accidents. But if God has no intrinsic accidents, two further implications follow. First, God cannot undergo real change, since he could only do so by gaining or losing intrinsic accidents. Second, God is intrinsically the same across all worlds in which he exists,

The Extrinsic Model Defended

93

since the only way that God could differ intrinsically in another world would be by having different intrinsic accidents in that world. If we assume, furthermore, that DUC is true, that God is a necessary being existing in all possible worlds and that God could bring about different effects than he does (or no effects at all), then it follows that there are other possible worlds and that God is intrinsically the same across all of them. It also follows that for some creaturely effect of God, E, there is a possible world in which God is exactly the same as he is in the actual world, but in which E does not exist. EM provides an account of divine agency that is consistent with the foregoing picture. For EM denies that God’s causing E involves some intrinsic feature of God that would be otherwise were God not causing E. Thus, EM is consistent with God’s being intrinsically the same across worlds where his effects differ. But herein resides a possible objection to EM. For, if E exists in the actual world, but God is intrinsically the same across worlds that do not include E, then nothing about God seems to make any difference to whether E exists. But, in that case, it may seem that God doesn’t account for E. And, if God doesn’t account for E, then the claim that God causes E may appear unintelligible. What is true of E is also true of that collection of creaturely effects we call “the universe.” The theist (and certainly the proponent of DUC) wants to say that God causes the universe. But, according to EM, God can be intrinsically the same in some possible world W*, in which there is no universe at all, as he is in the actual world W. In virtue of what, then, does God account for the universe’s actually existing? If God is the same across worlds, there doesn’t appear to be anything, and hence the claim that God causes the universe may appear without justification. Indeed, since there is no difference between W* and W prior to the universe’s existing, it appears that the universe just happens to be in W by chance, with nothing explaining or accounting for it.39 An objection very similar to the foregoing has been voiced by William Lane Craig. Craig’s specific target is Aquinas’s claim that, in creating, God is not really related to creatures: According to this [Aquinas’s] doctrine, then, God in freely creating the universe does not really do anything different than he would have, had he refrained from creating; the only difference is to be found in the universe itself: instead of God existing alone sans the universe we have instead a universe springing into being at the first moment of time possessing the property being created by God, even though God, for his part, bears no reciprocal relation to the universe made by him. I think it hardly needs to be said that Thomas’s solution, despite its daring and ingenuity, is extraordinarily implausible. “Creating” clearly describes a relation which is founded on something’s intrinsic properties concerning its causal activity, and therefore creating the world ought to be regarded as a real property acquired by God at the moment of creation. It seems unintelligible, if not contradictory, to say that one can have real effects without real causes. Yet this is precisely what Aquinas affirms with respect to God and the world. Moreover, it is the implication of Aquinas’s position that God is perfectly similar across possible worlds, the same even in worlds in which he refrains from creation as

94

Free Will and God’s Universal Causality: The Dual Sources Account in worlds in which he creates. For in none of these worlds does God have any relation to anything extra se. In all these worlds God never acts differently … he is just the simple, unrelated act of being. Even in worlds in which he does not create, his act of being, by which creation is produced, is no different in these otherwise empty worlds than in worlds chock-full of contingent beings of every order. Thomas’s doctrine thus makes it unintelligible why the universe exists rather than nothing.40

Clearly, certain elements of Craig’s criticism apply neither to EM nor to Aquinas. For instance, from God’s being intrinsically the same across worlds, it does not follow that God acts the same in these worlds. Such a charge presupposes that actions are, or involve, counterfactually varying intrinsic states of their agents, so that in order to act differently across worlds, God must himself differ intrinsically. EM rejects this presupposition, since, according to EM, God acts differently in different worlds just in case he bears causal relations to different effects in these worlds. That Aristotle and Aquinas reject this presupposition can be appreciated by recalling the discussion at the end of Section 5.1. Yet, even if one grants that EM allows God to act differently across worlds despite being himself intrinsically the same, the objection remains that it is unintelligible to claim God the cause of an effect when the way God exists prior to causing it is perfectly consistent with the effect’s not existing. This objection, which I will henceforth call the Intelligibility Objection, can be set out succinctly as follows: 1. If EM, then it is possible that for some creaturely effect of God E in the actual world W, there is a possible world W* in which God is intrinsically the same as he is in W prior to causing E, but in which E does not exist. 2. It is not possible that for any x and any y, if x causes y in the actual world W, then there is a possible world W* in which x is intrinsically the same as it is in W prior to causing y, but in which y does not exist. 3. Therefore, it is not the case that EM. The name “Intelligibility Objection” was chosen because the primary support for premise (2), as we have seen, is the claim that it makes no sense to say that x causally accounts for y, if x’s existing just as it is prior to causing y is consistent with y’s not existing.41 Whatever intuitive appeal premise (2) may have, however, I urge three reasons for rejecting it.42 First, premise (2) is incompatible with the most popular accounts of libertarian freedom. This point is especially significant given the dialectical context of my project. Since my purpose is to show that by accepting EM libertarians need not reject DUC, my project will be of greatest interest to those who think that libertarian freedom is coherent and plausible. To show that most libertarian accounts already reject premise (2) is, thus, to show that EM can be defended against the Intelligibility Objection without asking of my primary audience that they accept anything to which they are not already committed (or, at least, anything they do not already deem coherent and plausible).

The Extrinsic Model Defended

95

The reason that most libertarians are committed to rejecting premise (2) is that the most popular libertarian accounts include elements of indeterministic causation, whereby one entity x causes another entity y, without x’s necessitating or being logically sufficient for y. If, in the actual world W, x causes y without necessitating or being logically sufficient for y, then, contrary to premise (2), there is a possible world W* in which x is intrinsically the same as it is in W prior to causing y, but in which y does not exist. Agent-causal libertarian accounts clearly embrace indeterministic causation. On these accounts, an agent causes its act and/or the agent’s basic act consists in its causing some effect, usually a state of intention or a bodily movement. Since most libertarians want to say that an agent acting freely could at least sometimes have done otherwise all antecedent conditions remaining the same,43 it follows that the way such an agent exists prior to causing its act (or causing the effect in which its act consists) does not necessitate its act (or that effect). In other words, contrary to premise (2), when such an agent causes its effect in the actual world W, there is a possible world W*, in which that agent is intrinsically the same as it is in W prior to causing its effect, but in which its effect does not exist. Alongside agent-causal approaches, the most popular libertarian accounts are event-causal or reason-causal accounts.44 On these accounts, a free act is caused by certain agent-involving events, such as the agent’s reasons or motives, and the act either is undetermined by its proximate cause or there is indeterminism earlier in the causal process that results in the act. These accounts, thus, embrace causes that do not necessitate their effects. They therefore allow that a cause might have been the same, yet its effect not forthcoming, contrary to premise (2). Notice that despite the rejection of premise (2), in the case neither of God nor of the creaturely libertarian agent do we have an effect mysteriously appearing from nowhere. The creaturely libertarian agent-cause, for example, has the power to bring about its effect and has prior cognitive states (e.g., its deliberations or awareness of reasons) that are relevant to the alternatives it can bring about, even though these cognitive states do not determine the agent to cause any one alternative. By the same token, God has a prior cognitive state relevant to his bringing about whatever he can create—which state consists in God’s prior knowledge of all that it is possible for him to create and whatever reasons there might be for creating the different things he can. Of course, God’s power and prior cognitive state extend to many, perhaps infinitely many, more potential objects than do the creaturely libertarian agent’s. EM, then, accords with a traditional view on which God is free to choose between many, perhaps infinitely many, possible options for creation. Proponents of this view regarding God’s power and freedom gain nothing in response to the Intelligibility Objection by positing an intrinsic accident in God, such as God’s choice to create this particular universe, for the purposes of explaining why this universe exists rather than that. For this move only relocates the intelligibility question to why God has this particular intrinsic accident rather than any of the multitude of others he might have had, had he made a different choice. Better for a proponent of the traditional view of God’s power and freedom simply to reject the assumption that God can’t cause an effect, if God’s prior intrinsic state is consistent with God’s causing a multitude of alternative effects.

96

Free Will and God’s Universal Causality: The Dual Sources Account

A second reason for rejecting premise (2) of the Intelligibility Objection concerns the very concept of causation. As we have seen, premise (2) amounts to the claim that causes necessitate their effects. This claim is highly questionable. At her inaugural lecture at Cambridge, for instance, Elizabeth Anscombe famously maintained that the essential feature of the causal relationship is derivativeness, not necessitation: There is something to observe here, that lies under our noses. It is little attended to, and yet still so obvious as to seem trite. It is this: causality consists in the derivativeness of an effect from its causes. This is the core, the common feature, of causality in its various kinds. Effects derive from, arise out of, come of, their causes. … Causation, then, is not to be identified with determination. If A comes from B, this does not imply that every A-like thing comes from some B-like thing or set-up or that every B-like thing or set up has an A-like thing coming from it; or that given B, A had to come from it, or that given A, there had to be B for it to come from. Any of these may be true, but if any is, that will be an additional fact, not comprised of A’s coming from B.45

Anscombe’s contention seems quite plausible. Contrary to premise (2), for x to cause y, y must come from x, but y need not be necessitated by x. A third reason to reject premise (2) is that the premise has implications that are, in my view, less plausible than the negation of the premise. As we know, premise (2) entails that all causes necessitate their effects. But according to a basic modal truth, whatever entity necessitates a contingent entity is also contingent. From both the modal truth and premise (2), it follows that if a contingent entity has a cause, then that cause is contingent. Suppose, then, we believe (vii) that every contingent entity has a cause that brings it about. (vii) requires a cause of every contingent entity, and premise (2) together with the modal truth entails that the cause of every contingent entity will itself be contingent. From the modal truth, premise (2), and (vii), it thus follows that (ix) for any contingent entity, there is an infinite regress of causes resulting in that entity. The modal truth is, I think, unassailable. Thus, if we accept premise (2), we are committed either to rejecting (vii) or to accepting (ix). But rejecting (vii) or accepting (ix) is, in my view, less plausible than rejecting premise (2). I, thus, have reason to reject the premise, as will anyone who agrees with these judgments about plausibility. My defense of EM against the Intelligibility Objection has reached its conclusion. In my view, the objection should be set aside by rejecting premise (2). Premise (2) embodies a questionable understanding of the nature of causality, commits us to one of two less plausible claims than its rejection, and conflicts with the most popular accounts of libertarian agency. The difficulties libertarians face in accepting premise (2) are especially significant given the dialectical context of the current project. EM appeared vulnerable to the charge that it makes divine causality unintelligible, because it is plausible that God causes E only if God accounts for E. And it may seem that God doesn’t account for E, or make any difference to whether E exists, if the

The Extrinsic Model Defended

97

way God is intrinsically is consistent with E’s not existing. If, however, we reject the claim that causes must necessitate their effects, then we can allow that God causes E, even if the way God exists prior to doing so is compatible with E’s nonexistence. What’s more, if God causes E, then surely he accounts for E—a cause accounts for its effect. Finally, since that which an agent brings about intentionally is not a matter of chance, neither, given EM, is E’s existence (or the universe as a whole) a matter of chance, even though its nonexistence is consistent with God’s intrinsic state prior to causing it. In this chapter, I have presented and defended the extrinsic model of divine agency. In the following chapter, we will consider whether DUC and Dual Sources make God the cause of sin.

6

Does God Cause Sin?

The doctrine of divine universal causality (DUC) implies that God causes every creaturely action. We saw in Chapter 4 that, given Dual Sources, God’s causing creaturely actions is consistent with creaturely agents’ satisfying libertarian conditions for freedom and responsibility. Yet, even if creatures can be morally responsible for actions that are caused by God, a problem remains. Since some of our actions are sinful, God’s causing our actions may seem to imply that God causes sin. Drawing ideas from Anselm and Aquinas, this chapter shows, on the contrary, that God’s causing the act of sin does not entail that God causes sin itself. Only the creature, not God, causes sin.

6.1 DUC, Moral Evil, and the Privation Solution There is a simple argument from DUC to the conclusion that God causes sin: 1. God directly causes all creaturely actions (an implication of DUC). 2. Some creaturely actions are sins. 3. So, God directly causes sins. The logic of the argument may appear inescapable, and yet its conclusion is one that most theists have wanted to avoid.1 One reason for resisting the conclusion has been removed in light of the argument of Chapter 4. It might have been thought that if God causes a creature’s sin, then the creature cannot be responsible for the sin. Chapter 4 shows, on the contrary, that we can be responsible for our acts, even if God causes them. Nevertheless, reasons remain for resisting the conclusion that God causes sin. Sin is supposed to be something of which God disapproves in its own right, even if he sometimes permits sin for the sake of other goods; yet there is at least a tension between the claim that God disapproves of sin and the claim that God causes it. Similarly, sin is something for which many theists believe God sometimes punishes sinners, a belief that also appears in tension with the claim that God causes that very sin. Yet again, sin is a type of evil, and some have thought that that which is directly brought about by a cause reflects something of the nature or character of the cause.2 But, then, God’s directly causing sin would imply an aspect of evil within the divine nature. Finally, theists may reject the claim that God causes sin

100

Free Will and God’s Universal Causality: The Dual Sources Account

out of doctrinal or ecclesial commitments. For instance, the sixth canon concerning justification at the Council of Trent anathematizes the claim: “God works evil works just as he works good ones, not only by permitting them, but also properly and directly.”3 Although classical proponents of DUC such as Anselm and Aquinas clearly affirmed that God causes the act of sin, they denied that God causes sin, for they held that only the sinner, not God, causes the privation in virtue of which such acts are sinful. Thus, in On the Fall of the Devil, Anselm writes: Insofar as the will and its movement or turning are real they are good and come from God. But insofar as they are deprived of some justice they ought to have, they are not absolutely bad but bad in a sense, and what is bad in them does not come from the will of God or from God as he moves the will. Evil is injustice, which is only evil and evil is nothing. … Therefore, what is real is made by God and comes from him; what is nothing, that is evil, is caused by the guilty and comes from him.4

Similar to Anselm, Aquinas holds: God is the cause of every action, in so far as it is an action.—But sin denotes a being and an action with a defect: and this defect is from a created cause, viz. the free-will, as falling away from the order of the First Agent, viz. God. Consequently this defect is not reduced to God as its cause, but to the free will … Accordingly God is the cause of the act of sin: and yet He is not the cause of sin, because he does not cause the act to have a defect.5

In his reply to the second objection of the same article, Aquinas states his approach this way: Not only the act, but also the defect, is reduced to man as its cause, which defect consists in man not being subject to Whom he ought to be, although he does not intend this principally. Wherefore man is the cause of the sin: while God is cause of the act, in such a way, that nowise is He the cause of the defect accompanying the act, so that He is not the cause of the sin.6

What, then, are Anselm and Aquinas proposing?7 As I understand this privation solution, it involves the denial that any action, considered just in itself, is a sin. Rather, a sinful action consists of two elements: an act and a defect in virtue of which the act is sinful and in which the act’s sinfulness consists. We can certainly speak of bad or sinful acts, but since the acts are bad or sinful in virtue of defects distinct from (i.e., not identical to) the acts, to cause a “sin of action” requires causing both the act and the defect. As the universal cause, God causes the act and all its positive properties, since these are entities distinct from God. But the defect is not an entity and so is not something God must be said to cause just in virtue of DUC. Rather than an entity, the defect is a privation, a lack of something that should belong to the act. Anselm characterizes the privation as a lack of justice. In the

Does God Cause Sin?

101

passages cited above, Aquinas characterizes the defect as the act’s lack of proper order, or subjection, to God. In other places, Aquinas speaks of the act’s lack of conformity to the rule of reason or the divine law.8 For our purposes, these differences are not important. The essential and common point is that the privation in which the badness of a sinful act consists is the act’s lack of conformity to the moral standard, however exactly that standard is to be understood. The claim, then, is that while God causes the act and its positive properties, only the sinner causes both the act and the lack of conformity to the moral standard in which the act’s sinfulness consists. Thus, only the sinner, not God, causes the sin. Consider, again, the argument with which we began: 1. God causes all creaturely actions (an implication of DUC). 2. Some creaturely actions are sins. 3. So, God causes sins. We can now see how the privation solution enables a response. There is an ambiguity in the term “creaturely actions.” Since a sin of action consists of an act along with the act’s lack of conformity to the moral standard, premise (2) is true only if “creaturely actions” means actions along with their lacks of conformity to the standard. Yet, according to this meaning of “creaturely actions,” premise (1) is false, for God does not cause acts along with their lacks of conformity to the standard but only the acts themselves. Premise (1) is, thus, true only if “creaturely actions” means just creaturely acts, not including any lacks of conformity to the moral standard those acts may have. In short, there is no consistent meaning of “creaturely actions” on which both premises of the argument are true. And, of course, if the premises equivocate on the meaning of “creaturely actions,” then the argument commits a fallacy.9 Despite its promise to reconcile DUC with the denial that God causes sin,10 some may wonder whether the privation solution is even consistent. For DUC claims that God causes all entities distinct from himself. Yet the solution, though it denies that God causes the privation in a sin of action, appears to speak of the privation as if it were an entity. For instance, just above, the privation was spoken of as one of two elements or constituents in a sin of action; and it was spoken of as caused by the sinner. Aren’t constituents of things, and objects of causation, entities? And if they are, won’t the proponent of DUC be required to say that God causes the privation after all? Appearances notwithstanding, a proponent of the solution need not admit that privations are entities. Consider Aquinas’s distinction, borrowed from Aristotle, between two senses of “to be”: Note then that Aristotle says there are two proper uses of the term being: firstly, generally for whatever falls into one of Aristotle’s ten basic categories of thing, and secondly, for whatever makes a proposition true. These differ: in the second sense anything we can express in an affirmative proposition, however unreal, is said to be; in this sense lacks and absences are, since we say that absences are opposed to presences, and blindness exists in an eye. But in the first sense only what is real is, so that in this sense blindness and such are not beings.11

102

Free Will and God’s Universal Causality: The Dual Sources Account

We speak truly when we say that privations, like blindness, exist. But that does not make privations real; it does not make them entities, the sort of things that fall within Aristotle’s categories or within the scope of what DUC says God causes. To say that a privation exists is not to say that there is something real there, an entity, but rather that what should be there is missing.12 Now, any human act issuing from reason and will should conform to the moral standard. If such an act does not so conform, then what should be there is missing. And, since, according to the privation solution, an act’s sinfulness consists precisely in its lack of conformity to the moral standard, we can say that this lack is an “element” within the sin of action. But that doesn’t imply that the lack of conformity is a real entity like the act itself. Similarly, we can speak of the cause of a privation—as when we say that a man’s blindness was caused by too much sunlight—without thereby committing ourselves to the claim that the privation, like the man himself, is an entity. A cause of a privation is just whatever is explanatorily responsible for the fact that the deprived thing we are talking about lacks what it ought to have. Even if we put this initial worry about consistency aside, there remain formidable objections to the privation solution. First, the solution depends on a privation account of moral evil, according to which the badness of a morally bad act consists not in the positive act or any of its positive properties but rather in a privation of conformity to the moral standard. While such an account has its contemporary defenders,13 it also has a number of critics whose objections must be answered for a defense of the solution.14 Second, it may be thought that, given DUC, even if one allows that the badness in a sin of action consists in a lack of conformity to the moral standard, God will still be a cause of this lack of conformity simply by causing the act and its positive properties. After all, it is because the act and its positive features are what they are that the act fails to conform to the standard: The lack of conformity would seem simply to follow on the act and its positive features. And, of course, if God is the cause of the privation as well as the act and its positive features, the privation solution fails. Although this objection has been surprisingly neglected by contemporary philosophers, it had an able spokesman in the early Leibniz who charged that “the privation is nothing but a simple result or infallible consequence of the positive aspect”15 and that “it would be a joke to say that [someone] is the author of everything real without … being the author of the privative aspect.”16 Finally, a third objection holds that appeal to privation as a way of blocking the claim that God causes sin will make it impossible to affirm that the sinner causes sin. For if God can cause the act and its positive features without causing the privation, won’t the same be true of the sinner? As Leibniz puts it, “I am amazed these people did not go further and try to persuade us that man himself is not the author of sin, since he is only the author of the physical or real aspect, the privation being something for which there is no author.”17 Of course, proponents of the privation solution insist that the privation does have a cause, namely, the sinner. But how? As Samuel Newlands asks, “If God does not cause absences, how can we? And if we can, why cannot God?”18 Without an answer to these questions, the privation solution appears feeble, indeed.

Does God Cause Sin?

103

In the remainder of this chapter, I defend the privation solution against these objections. In Section 6.2, I offer some positive philosophical support for the privation account of moral evil, showing that a case can be made for the account distinct from a case based on its theological usefulness. In Section 6.3, I then respond to the most important objections to the privation account of moral evil. In Section 6.4, I argue that, in fact, God does not cause a sinful act’s lack of conformity to the moral standard simply by causing the act and its positive features. And in Section 6.5, I explain how we can hold the sinner causally responsible for his act’s lack of conformity to the moral standard, even while denying that God is causally responsible. The privation solution, it turns out, is a viable approach to reconciling God’s universal causality with the denial that God causes sin.

6.2 Moral Evil and Privation In this section, I offer a positive philosophical argument in support of the privation account of moral evil. The argument admittedly depends on substantive and controversial positions in metaphysics, action theory, and ethics, positions generally embraced within the scholastic tradition, but for which it will not be possible to provide a complete defense in the space available. For this reason I do not claim to demonstrate the truth of the privation account of moral evil. But I do offer enough to indicate that a serious philosophical case can be made for the account, even apart from a case based on its theological usefulness. The argument for the privation account comes in two stages or sub-arguments. The first is an argument aiming to show that the badness of a sinful act consists in something distinct from the exercise of active power.19 This conclusion is significant, for when the proponent of the privation account denies that an action, considered just in itself, is a sin, what he means by an “action” (I take it) is a particular “exercise of active power.” If the badness of a bad action is something distinct from the exercise of active power, then something in addition to an act will be needed to make a sin. The second stage proposes “lack of conformity to the moral standard” as the best candidate, both for that which (in addition to an exercise of active power) is needed to constitute a sinful act and as that element in which the sinfulness of such an act consists. Stage One. The argument comprising the first step goes as follows: 1. All actions are exercises of active power (assumption). 2. All exercises of active power are, qua exercises of active power, good (assumption). 3. So, all actions are, qua exercises of active power, good (inference from 1 and 2). 4. If (3), then no action, qua exercise of active power, is bad (assumption). 5. So, no action, qua exercise of active power, is bad (inference from 3 and 4). 6. If (5), then, if there are bad actions, then the badness of bad actions consists in something distinct from the actions qua exercises of active power (assumption). 7. There are bad actions (assumption). 8. Therefore, the badness of bad actions consists in something distinct from the actions qua exercises of active power (inference from 5–7).

104

Free Will and God’s Universal Causality: The Dual Sources Account

Assumption (1). The actions included within the scope of this premise (and of the argument as a whole) are not limited to the deliberate actions of rational agents but rather include the actions of any substance or integral part of a substance. Thus, a trout’s swimming, a heart’s pumping blood, and a person’s signing her name all count as actions for the purposes of this argument. The claim is that all such actions are exercises of an active power (or powers), belonging to an agent. For something to act is for it to exercise active power. Assumption (2). Even though this is probably the most controversial assumption of the argument, I believe the assumption plausible. An active power is the power to do some particular type of thing. The power of sight is the power to see things. The power of speech is the power to speak. The power of reason is the power to reason. That which a given power is a power to do is the “end” or “goal” of the power. It is something to which the power is ordered or directed, i.e., that at which the power aims. The good of a power is the achievement of its aim, the doing of that which the power is a power to do. Thus, when the eyes see and the power of sight is exercised, the power of sight achieves its aim. Such achievement is good. So, an exercise of active power is as such something good.20 Notice that an active power may be exercised well or poorly, depending on how well it does what it is a power to do. I have the power of sight, but my eyes (at least unaided) do not see well. If they did not see at all, we could not even say that the power of sight had been exercised. They see to some extent, and so the power of sight is exercised, and to that extent the power of sight achieves its aim; the exercise is good. And yet my eyes see poorly or defectively. Even though they achieve something of the aim to which the power of sight is directed, my seeing falls short of the measure of healthy seeing. I do not see all that I should. My seeing is bad, then. But the badness of my seeing consists in its falling short of the standard for healthy seeing, not in the exercise of my power of sight, in which I do to some extent see, and thus achieve something of the aim of the power, which is good. Now, consider a sin of action, say, Smith’s robbing a convenience store for the sake of obtaining money to purchase a car. Various active powers are exercised in this act of theft, as in most sins of action. By the power of will, Smith desires money to purchase a car and chooses to rob a convenience store in order to obtain the money. Smith’s intellect or reason is operative in presenting the car as something desirable, in judging that sufficient money would enable Smith to purchase a car, and in recognizing robbing the convenience store as a means of obtaining the money. Various powers of Smith’s body are employed, as directed by Smith’s reason and will, in pulling the robbery off. The act is bad, no doubt. It is morally bad. Yet the various powers exercised in the act are, in that exercise, achieving to some extent what they are powers for. The will is a power, among other things, for desiring something under a respect in which it is, or seems to be, good; for choosing a means to obtain the good desired; and for moving other powers, including powers of the body, toward the execution of the means. Reason or intellect is a power, among other things, for noticing respects in which various things are good and for proposing effective means for the obtaining of goods desired. The body’s physical powers are, among other things, for affecting the physical environment as directed by reason and will. In Smith’s act of theft, Smith’s

Does God Cause Sin?

105

reason, will, and bodily powers are exercised in all of these ways, thereby achieving something of what those powers are for. Indeed, as with the example of sight above, if these powers did not achieve something of what they are for, we could not even say that the powers had been exercised. Since such achievement is good, the exercise of these powers is as such good, with the consequence that Smith’s act qua exercise of active power is good. It is important to see that claiming that exercises of active power are as such good does not entail that these exercises cannot be deemed bad when considered in relation to other things. When the jaws of a lioness bear down on the neck of a lamb, the exercise of the lioness’s powers is clearly bad from the standpoint of the lamb, even though it is an achievement of those powers and, hence, good. Similarly, the exercises of the various active powers involved in the theft described above are qua exercises good, since they are, at least to some extent, achievements of those powers’ aims. But if the theft is considered relative to the good of Smith as a whole, it would have been better had he not exercised those powers in the way that he did. For the good for a human being requires that his acts conform to the moral standard. An act of theft does not so conform. One can then affirm that all exercises of active power are as such good, without thereby denying that there are respects in which such exercises are bad. As the argument goes on to show, to say that the exercises of active power are as such good only requires that if they are also in some way bad, they are bad in virtue of something distinct from the exercise of active power. It is worth adding that even those unfamiliar with arguments of the sort given here often witness to the truth of the conclusion when they recognize, for instance, in the “criminal mastermind” or “supervillain” impressive achievement in the exercise of power even in the commission of morally bad acts. There is a certain type of weakminded person that can become so taken with such achievement that he actually begins to admire the master criminal, despite the fact that the criminal is doing something morally bad. Assumption (4). Good and bad are contraries. Thus, if something is good qua x it cannot also be bad qua x. If something is good and bad, that on account of which it is bad will be distinct from that on account of which it is good. Assumption (6). This assumption is self-explanatory. If no action is bad qua exercise of active power and yet if some actions are bad, then the badness of those actions must consist in something other than the actions qua exercises of active power. Assumption (7). If there were no bad actions, then the problems of this chapter would not even arise. That there are bad actions is a central assumption of the whole discussion. Conclusion (8). The conclusion of this first stage in support of the privation account of moral evil is that the badness of bad actions consists in something distinct from the actions qua exercises of active power. As noted above, this conclusion is significant because an exercise of active power is what proponents of the privation account of moral evil understand an “action” to be when they deny that an action, considered just in itself, is a sin. Sins are bad, but if the conclusion at (8) is correct, then that badness is found in something distinct from the action, understood as an exercise of active power. As established at inference (5), no action qua exercise of active power is bad.

106

Free Will and God’s Universal Causality: The Dual Sources Account

Stage Two. In what, then, does the badness or sinfulness of a sinful act consist? With this question we have reached the second stage in the support for the privation account of moral evil. We know from the first stage that the badness has to be something distinct from the act identified as an exercise of active power. Could it be that the badness consists in some positive feature or property distinct from yet belonging to the act? But at least as suitable a candidate is suggested by the discussion of my defective vision under my defense of assumption (2) above. There I noted that the badness of my seeing consists not in the exercise of my power of sight but rather in my seeing’s falling short of the standard for healthy seeing. This falling short is not a positive feature or property.21 Rather, it is a privation, a lack of something that my seeing should have, namely, conformity to the standard. The privation account of moral evil proposes a similar analysis of the badness of sins of action. That badness consists not in the exercise of active power but in the act’s or exercise’s lack of conformity to the moral standard. Human acts—acts over which the human has mastery by reason and will22—ought to conform to the moral standard. If they do not so conform, they are bad. But, as in the case of my defective seeing, the badness consists not in the exercise of active power but in the privation, the lack of conformity to the standard. The badness of Smith’s theft lies in its injustice, its failure to agree with the principles of justice, one of which is not to take what belongs to another. Given that the badness of a sinful act is something distinct from the action qua exercise of active power, the most obvious alternative to the privation account of moral evil is the one briefly suggested above, namely, to identify the badness of a sin of action with some positive property or properties distinct from, and yet belonging to, the act. What property or properties? Here I consider what seem to be the three most likely candidates. First, it might be proposed that all bad things, bad human acts included, have a “badness” property. One initial attraction of this view is that it assigns a property (or property-type) in common to all that we call “bad”; the “badness” of all bad things refers to a single “badness” property (or property-type) had in common by those things. Whatever its attractions, serious doubt has been cast on this proposal by the observation of Peter Geach and others that “good” and “bad” are attributive rather than predicative adjectives.23 Whereas the content of a predicative adjective such as “red” remains the same regardless of the noun that it modifies, the content of an attributive adjective does not. We know what specifically it is for X to be red even without knowing what X is. But we do not know what specifically it is for X to be good or bad without knowing what X is. The goodness of a pillow is quite different from the goodness of a hammer. The badness of an ankle is not the same as the badness of an argument. That goodness and badness are not specifically the same in all things testifies against the proposal that the badness of all bad things, including bad human actions, consists in their sharing a common badness property (or property-type). A second candidate for a property in which the badness of an action consists is the property of “being opposed to goodness.” On one understanding of “opposed to goodness” this proposal surely fails. For, if “opposed to goodness” just means that the act harms certain goods, then there are many acts opposed to goodness that are not necessarily bad or sinful. Firing an employee is opposed to the goodness of

Does God Cause Sin?

107

the employee’s livelihood. Imprisoning a criminal is opposed to the goodness of the criminal’s liberty or his ability to support his family. Killing in just self-defense or war is opposed to the goodness of the life of the one killed. Since acts can harm goods without being bad or sinful, it will not do to say that the badness of an action consists in its being “opposed to goodness” in this sense. Alternatively, by “being opposed to goodness” one might mean “conflicting with moral goodness.” Any act that had the property of “being opposed to goodness” in this sense would be bad or sinful. Having the property of “being opposed to goodness” in this sense would be tantamount to having the property of “conflicting with the moral standard.” While the property, so understood, avoids the problem with the first understanding of a “being opposed to goodness” property, I do not see any advantage that identifying the badness in a sin of action with a positive property of “conflicting with the moral standard” has over identifying the badness with a lack of conformity to the moral standard. Thus, there would not appear to be any reason to prefer this approach to the privation account. On the other hand, there is reason, even independent of theological concerns, to prefer the privation account to any alternative that understands the badness of an action to consist in a positive property. For, while I have not explicitly defended the doctrine that all evil is privation, one advantage of the privation account of moral evil is that it is consistent with this doctrine.24 The claim that all evil is privation is attractive because it is the natural companion to an attractive conception of goodness. Above we noted that “goodness” is not specifically the same for all good things; the goodness of a pillow differs from the goodness of a hammer. We should thus deny that the “goodness” of all good things consists in a property (or property-type) that they share in common. Nevertheless, it seems false to say that the term “good” applied to a pillow and a hammer is used purely equivocally, without any common meaning. An appealing solution is to say that an individual of type X is good to the extent that it fulfills the standards for an X and so has whatever an X ought to have. We might add that something Y (e.g., a property, relation, or activity) is good to the extent that it is an achievement, actualization, or perfection of an individual, given its type.25 This solution affords the term “good” some common meaning when we predicate it of a pillow and a hammer—each has what it ought to have given the kind of thing that it is—while also denying that the goodness of a pillow and the goodness of a hammer consist in the same characteristic(s). The solution also suggests a privation account of badness or evil. If something is good to the extent that it has what a member of its type ought to have, then, very plausibly, it is bad to the extent that it lacks what it ought to have. Its goodness consists in its having what it ought. Its badness consists in lacking what it ought. That the privation account of moral evil is consistent with a general privation account of evil speaks in its favor.26 Finally, as a third candidate for a positive property or properties with which the badness of a sinful act could be identified, one might propose the properties in respect of which various sinful acts conflict with, or lack conformity to, the moral standard. For example, maybe an act has the property of being done for the sake of harming someone and so conflicts with the standard in respect of that property. Maybe another

108

Free Will and God’s Universal Causality: The Dual Sources Account

act has the property of being boastful and so conflicts with the standard in light of that property. On this view, the badness in sinful acts will vary depending on the act. It will consist in whatever properties of the act are the properties in respect of which it conflicts with the moral standard. Let us grant for the sake of argument that there are such properties. There are still at least two reasons, even apart from theological concerns, for adopting the view that the badness in sins of action consists not in these properties but in the lack of conformity to the moral standard. First, whatever plausibility accrues to identifying the badness in sinful acts with these properties derives from the fact that these properties are those in respect of which an act lacks conformity to the moral standard. But the fundamental reason the act is morally bad is that it departs from the standard. It may be true to say that it is morally bad because it has these properties, but only because acts that have these properties lack conformity to the standard. If acts that had these properties did not conflict with the moral standard, then they would not be morally bad. Since the fundamental reason an act is morally bad is its departure from the moral standard, it is more fitting to identify the badness with the lack of conformity itself than with the properties in respect of which it lacks conformity.27 The second reason for preferring the view that the badness consists in the act’s lack of conformity to the standard is, again, that this view is consistent with a privation account of evil in general, an account that is philosophically attractive for the reasons discussed above. We have seen that there is good reason, even apart from theological concerns, for embracing the privation account of moral evil. That having been established, it would not be unreasonable to embrace the privation account, even if the only reason for preferring it was its usefulness in solving the theological problem with which we started. Provided that the alternative to the privation account is not obviously preferable on other grounds and that the privation account can be defended against objections, the fact that the account makes possible a resolution of the theological problem is an excellent reason to endorse it. To a consideration of objections to the privation account I now turn.

6.3 Objections to the Privation Account of Moral Evil In this section, I respond to what seem to me the three most formidable objections to the privation account of moral evil.28 Objection 1. At least some sins of action cannot be adequately characterized on the privation account. Todd Calder has recently pressed this objection. He writes: In some cases evil is not just the absence of goodness but rather some positively bad existing property or quality … For instance, the malicious torturer is not just not as good as she might be. She is not simply withholding gestures of kindness which a morally decent person would bestow; her actions are positively bad and these actions are constituted by attributes she possesses, i.e., desires for other people’s pain for pleasure, and not by attributes she lacks.29

Does God Cause Sin?

109

Later Calder states: To adequately characterize murder we must say more than that murder is the nonfulfillment of the duty to refrain from unjustified killing: we must say what it is to fail to refrain from unjustified killing. To do so, we must say what it is to kill without justification. But we cannot say what it is to kill without justification without reference to attributes possessed by the murderer, such as her beliefs, desires and intentions. But if our analysis makes reference to attributes possessed by the murderer, we no longer have a privation theory of evil.30

These statements, culminating in the last sentence of the second passage, indicate that Calder either ignores or misconstrues the privation account as found in its classical proponents, such as Anselm and Aquinas. Contrary to Calder’s suggestion, the proponent of the account does not think that murder can be adequately characterized simply as the nonfulfillment of a duty. Murder, as with all sinful acts, involves a positive act (and whatever positive attributes this implies) and not just a privation of conformity to the moral standard. To admit as much means “we no longer have a privation theory of evil” only on an idiosyncratic understanding of the theory. Moreover, the proponent of the privation account, as I have presented it, can agree that the malicious torturer’s (and the murderer’s) acts are something positive and that they are bad. The privation theorist is not claiming that no positive entity can be denominated “bad.” The claim is rather that the badness of a positive entity, that in virtue of which it is denominated “bad,” is not itself a positive entity but rather a privation. To alter slightly Calder’s example in the first passage, if a person wills someone’s pain for the sake of pleasure, this act is a bad positive entity. The privation theorist insists only that the badness of the act consists not in the act as positive entity but rather in the act’s lack of conformity to the moral standard. Calder offers nothing to refute the privation theorist on this point.31 Objection 2. The privation account cannot explain why certain sins of commission are worse than certain sins of omission. This objection has been voiced by Stanley Kane, who writes: There is a familiar distinction, and one of great moral importance, between the failure to perform loving acts on the one hand and the performance of hateful or murderous acts on the other—between sins of omission and sins of commission. Yet on the privation theory we would have to say that both sorts of sin are equally evil and that as evil there is really nothing in the hateful or murderous acts beyond the lack or privation of love and right action. This, it seems to me, is a reductio ad absurdum of the theory.32

Calder offers a similar objection in the course of criticizing Anglin and Goetz’s response to Kane: Anglin and Goetz argue that moral evils are privations as well … For instance, the evils of murder and of letting starving people in distant countries die consist in

110

Free Will and God’s Universal Causality: The Dual Sources Account

the nonfulfillment of the duties to respect and preserve life. According to Anglin and Goetz, murder is more evil than letting starving people in distant countries die since it constitutes a greater nonfulfillment of the duty to respect and preserve life. But how is murder a greater nonfulfillment of the duty to respect and preserve life? Sadly, Anglin and Goetz provide no answer to this question. It seems that the only way to account for why murder is a greater nonfulfillment of the duty to respect and preserve life is to refer to the intrinsic disvalue of attributes possessed by the murderer such as her intention and desire to take someone’s life without justification. But these are attributes she possesses, not privations, and thus it seems that even if we describe murder as the nonfulfillment of the duty to respect and preserve life, to distinguish it from other, less evil, forms of failing to fulfill this duty, we cannot characterize it solely in terms of the privation of good properties or attributes.33

Both Kane and Calder here fall into the mistake of Objection (1), assuming that on a privation account of moral evil, all moral evil must be characterized in a way that makes no reference to positive attributes or actions. In fact, this mistake appears to be what leads them to think that the privation account cannot distinguish the severity of a positive act of murder from the severity of neglecting to give the assistance needed to preserve the lives of those in a distant country. If no reference can be made to the positive elements in an act of murder (so the objection goes), we will not be able to articulate a specific moral rule to which the murderer fails to conform that is different from the rule to which the neglectful fail to conform. Both will have simply failed to abide by something such as the very general principle to respect and preserve life. It will not be possible to distinguish the murder from the neglect in the way required to say that the one is worse than the other. This problem disappears once one recognizes that, on the privation account here defended, a sin of action includes a positive act as well as a lack of conformity to the moral standard. Because we can make reference to the positive act, we can articulate a moral rule violated by the murderer that is distinct from the rule violated by the neglectful. The murderer fails to conform to the prohibition against intentionally killing the innocent, whereas the neglectful fails to abide by the positive norm to help those in serious need when one is able (taking into consideration all one’s obligations, etc.). Having distinguished the rules violated, we can coherently say that violating the rule against murder is (at least in most cases) worse than violating the precept to help those in serious need. We can even refer to aspects of the positive act to explain why murder is generally worse. For instance, murder involves aiming to kill, whereas neglect, even if it results in death, does not typically aim at the deaths of those neglected. We might add that the one is worse than the other since aiming to kill represents a greater failure to respect justice and the good of human life; that is, it represents a greater departure from the moral standard. All of this is, as far as I can tell, perfectly consistent with the privation theorist’s claim that the sinfulness in an act of murder consists not in the act qua exercise of active power but in the act’s lack of conformity to the moral standard. Objection 3. Some sins of action are opposed to goods in a way not consistent with the claim that their principle of badness is a mere privation. John Crosby writes:

Does God Cause Sin?

111

I think I speak for the moral common sense of mankind when I say that what makes Cain’s murder evil includes indeed, but is not restricted to, the lack of the brotherly love due to Abel. How can we fail to see that what makes it evil is also its turning against the good of Abel’s life and against the divine goodness shown to Abel? … How can anyone reasonably deny that good and evil are here related to each other, not as having and lacking, but as strictly contrary opposites?34

A bit later, Crosby continues: The act of Cain means to be the agent that reduces Abel from being to non-being. It is aggressively directed against Abel in such a way that the principle of its badness hardly seems to be some mere lack. When Cornwall puts out Gloucester’s eyes (King Lear), his cruel and violent act does not have the same merely privative character that the blindness he causes in Gloucester has.35

As should be clear by now, the privation account does not hold that Cornwall’s cruel act has the same merely privative character as Gloucester’s blindness. For Cornwall’s sin, unlike Gloucester’s blindness, includes a positive entity, an act. This correction notwithstanding, there is a sense in which what Crosby maintains in these passages is true, although it is not a sense that contradicts the privation account as I have presented it. As I have presented it, the sinfulness or badness of a sinful act consists in its lack of conformity to the moral standard. Now suppose the moral standard requires respect for certain goods in certain ways or under certain circumstances. And suppose that acts opposed to certain goods in the relevant ways or circumstances—as, for example, murder is opposed to the good of human life, when it is innocent—fail to respect these goods. Given these suppositions, an act may lack conformity to the moral standard for the reason that it is opposed to a relevant good in the relevant way. Since the reason that such an act lacks conformity to the moral standard is its opposition to certain goods and since the badness of the act consists in its lack of conformity to the standard, there is a legitimate sense in which the act’s opposition to certain goods is a principle of its badness. A proponent of the privation account can thus agree with Crosby’s point that opposition to good is a principle of badness in actions. Of course, the privation account would be undermined if we said that the badness of the act consisted not in its lack of conformity to the standard but in the act as positive entity or in some positive property of the act. So, the question arises: If we agree that Cain’s act lacks conformity to the moral standard because it is opposed to the good of Abel’s life, should we also conclude that the badness of Cain’s act consists in the act itself, which is so opposed, or in some positive property of “opposition to Abel’s life” that belongs to the act? I do not think so, and for reasons similar to those already discussed in Section 6.2. The fundamental reason why an act is morally bad is that it lacks conformity to the moral standard. If one claims that an act is morally bad because it is opposed to certain goods, the claim may be true, but only because an act that is opposed to these goods lacks conformity to the standard that demands respect for these goods. If an act opposed to these goods did not conflict with the moral standard, then it would not be morally bad. For these reasons, an act’s badness is more fittingly

112

Free Will and God’s Universal Causality: The Dual Sources Account

identified with the lack of conformity than with the act itself or some positive property of the act, in respect of which it lacks conformity. To be clear, it is perfectly consistent with the privation account to say that Cain’s act is morally bad or even that the act’s opposition to the good of Abel’s life is morally bad. But the act is bad because it lacks conformity to the moral standard, and the opposition to Abel’s life is bad because it is that in respect of which the act lacks conformity. Thus, the badness of the act and its opposition to Abel’s life consist not in the positive act or property but in the lack of conformity in virtue of which these items are bad.36 Having answered the most serious objections to the privation account of moral evil, let us turn to consider another set of objections to the privation solution.

6.4 Does God Cause the Badness in Sinful Acts Simply by Causing the Acts? According to the privation solution, the badness or sinfulness of a sin of action consists in its lack of conformity to the moral standard. This is a crucial part of the explanation of how the sinner alone causes sin, even though both the sinner and God cause the act of sin. Yet, in order for the privation solution to succeed, it must be shown further that unlike the act itself, the act’s privation of conformity to the moral standard is caused by the sinner alone, and not also by God. This may seem a difficult task. For, even though a lack of conformity is not an entity, and hence not something God causes as an implication of DUC, one might think with Leibniz that God causes an act of sin’s lack of conformity simply by causing the act and its positive properties. After all, the moral standard being what it is, it is not possible, say, for an “intentional killing of the innocent” to exist without lacking conformity to the standard. Since, given the standard, the lack of conformity would appear simply to follow from the act, won’t God cause the lack of conformity in causing the act? Moreover, as noted in the previous section, the privation account of moral evil allows that acts lack conformity to the moral standard in virtue of or in respect to certain positive properties, such as the property of aiming to kill the innocent. A proponent of the privation solution, committed to DUC, has to say that God causes these positive properties, insofar as they are entities distinct from God. But if God causes whatever positive properties are those in virtue of which an act lacks conformity to the moral standard, won’t God cause the lack of conformity, as well? An understandable objection, but I believe it can be resisted. Observe that in order to get an act’s lack of conformity to the moral standard, we need more than the act and the properties or features in respect of which the act lacks conformity. We also need the standard itself. It takes both the act with its positive features and the moral standard in order for the act to lack conformity to the moral standard. But this point suggests that something does not cause the lack of conformity to the standard simply by causing the act and its positive features. To cause the act and its positive features is not enough. The foregoing reasoning is an application of a more general point. To get a relation R between two relata a and b, or the lack of a relation R between a and b, or the truth of a relational proposition “aRb,” one needs both relata, a and b, and all their relevant

Does God Cause Sin?

113

properties. But, for this reason, a cause C cannot plausibly be thought to cause R, or the lack of R, or the truth of “aRb” simply by causing one of these relata and its relevant properties.37 Suppose, for example, that Cecilia makes a sandwich and Elizabeth makes an omelet and Elizabeth’s omelet weighs more than Cecilia’s sandwich. It is, of course, true that Elizabeth has made an omelet that weighs however many ounces. And it is true that she has made an omelet that weighs more than Cecilia’s sandwich. But has Elizabeth, simply by causing her omelet and its relevant properties, caused her omelet’s being heavier than Cecilia’s sandwich (or caused the truth of “Elizabeth’s omelet is heavier than Cecilia’s sandwich”)? I think not. Nor, if a spider spins a web and a robin builds a nest, and the spider’s web is more beautiful than the robin’s nest, does the spider cause its web’s being more beautiful (or the truth of “The web is more beautiful than the nest”). The reason for these negative judgments is that to get the relations or relational truths, you need both relata and their relevant properties. So, it is implausible to think that Elizabeth or the spider causes the relations or truths simply by causing one of the relata and its properties. Notice that it is not significant to the judgments about Elizabeth and the spider that the relata they don’t cause (the sandwich and the nest), and their properties, are caused by something else (Cecilia and the Robin). The spider’s web is less beautiful than God. God and God’s unmatchable beauty has no cause. But the spider no more causes its web’s being less beautiful than God than it causes its web’s being more beautiful than the nest. And the reason is the same. All the spider has done is cause half of what gives rise to the relation. Nor would it seem to make any difference if we add that Elizabeth knows that the omelet she is making is heavier than Cecilia’s sandwich. For that knowledge does not make her any more responsible for the sandwich and its properties, which are needed every bit as much as the omelet, in order to give rise to the relation. Nor would it matter even if we said that Elizabeth made her omelet a certain weight in order that it be heavier than the sandwich. For, while her goal certainly explains why she made it the weight she did, that goal, together with her making the omelet the given weight, does not bring about the omelet’s actually being heavier, since that relation depends also and immediately on the sandwich and its weight, which Elizabeth plays no role in bringing about. Perhaps, Elizabeth could plausibly be thought to cause the omelet’s being heavier than the sandwich, if she had made the sandwich as well as the omelet. In such a case, she would have caused all of that on which the relation immediately follows. By the same token, it arguably would be enough for God to cause the lack of conformity to the moral standard if God caused not only the act of sin and its positive features but also the moral standard to which the act fails to conform. But very many theists deny that God causes the moral standard, even if they believe that God shapes the content of that standard in certain ways. For even if a theist thinks, for example, that God brings it about that intentionally killing the innocent is wrong through a command, that command constitutes a moral standard for us only on the supposition that we must abide by God’s commands. And it would be a very radical divine command theory, which held that this most general norm to abide by God’s commands is brought about only by God’s commanding it. Suppose, alternatively, a moral theory

114

Free Will and God’s Universal Causality: The Dual Sources Account

that understands what we ought to do (or refrain from) to be determined by what’s required to flourish given the nature we have. A theist may reasonably deny that the content of human nature is caused by God; human nature may be an idea that God has from all eternity prior to any causal act on his part. And even if a theist holds that the content of human nature is brought about by God, God will not have caused the moral standard unless God also brings about the truth of the very general principle that what a thing ought to do is determined by its nature, a claim that proponents of this sort of theory may deny. In short, very many theists will deny that God causes the moral standard. But, given this denial, just as God doesn’t cause the lack of conformity to the standard simply by causing the act and its positive features, neither does he cause the lack of conformity by causing the act and its features along with the standard. The foregoing seems a welcome result. But did it prove too much? The privation solution requires relieving God of causal responsibility for the act’s lack of conformity to the moral standard, but it also requires that the lack of conformity be imputable to the sinner. But, presumably, the sinner no more causes the moral standard than God does. So, if causing the lack of conformity requires causing the moral standard, then the sinner does not cause the lack of conformity either, and the privation solution fails. Can one plausibly deny that God causes the lack of conformity without at the same time making it impossible to affirm that the sinner causes it? Are there grounds for imputing the lack of conformity to the sinner and only the sinner?

6.5 How the Badness in Sinful Acts Is Caused by the Sinner Alone I have argued that God does not cause a sinful act’s lack of conformity to the moral standard simply by causing the act and its positive features. While this conclusion is necessary for preserving the privation solution, it also raises a question about the basis for our affirming that the sinner causes sin. For, if the simple fact that God causes the act and its positive features is not enough to make God cause of the lack of conformity, neither is the lack of conformity imputable to the sinner simply from the sinner’s causing the act and its features. Moreover, holding that the sinner causes the moral standard will likely seem even less attractive to theists than holding that God causes the standard. Thus, it won’t do to say that the sinner causes his act’s lack of conformity by causing the moral standard along with his act. Yet, I have not claimed that causing the lack of conformity requires causing the moral standard. What I have argued is that to cause the lack of conformity, it is not enough simply to cause the act and its positive features. Causing the moral standard in addition to the act and its features would seem the most obvious and straightforward way for something to cause the lack of conformity. But there is another way. There is a way in which an agent might cause or account for a lack of conformity to a standard despite not causing the standard. In what follows, I suggest an explanatory framework according to which, indeed, the sinner causes the lack of conformity, but God does not. Let us begin by considering some examples.

Does God Cause Sin?

115

Suppose I make a kite for my daughters. The kite doesn’t conform to FAA standards for commercial airliners. I built a kite that lacks conformity to FAA standards, and I am well aware of this fact. But have I caused the kite’s lack of conformity to the standards? Have I caused the truth of “The kite doesn’t conform to FAA standards”? No more, I think, than Elizabeth has caused her omelet’s being heavier than Cecilia’s sandwich. Suppose, similarly, that I make a sled for my daughters to go sleigh riding. The sled lacks conformity to the standards for Olympic bobsleds. I have built a sled that lacks conformity to Olympic standards, but I have no more caused the sled’s lack of conformity than I cause the kite’s lack of conformity in the example above. But now suppose I have been hired to build a sled for use in the Olympic bobsled competition. And suppose, again, that I build a sled that lacks conformity to Olympic standards. Although I did not cause those standards, I am, this time, responsible for my sled’s lack of conformity to them. The lack of conformity is imputable to me, because, unlike before, I ought to have built a sled that conforms to those standards. I account for my sled’s lack of conformity in virtue of my having neglected to build according to the standards to which I was responsible. Notice that my claim here is not simply that this time, but not before, I am at fault for having built a sled that lacks conformity to Olympic standards. That much is true, but I want to claim further that this time, but not before, there is a way in which I, or my negligence, accounts for the lack of conformity. The fact that I have an obligation to build according to the standards means that this time, but not before, I have responsibility for whether the sled conforms. When the sled does not conform, the lack of conformity is, thus, accounted for by my not having built a sled of the sort I was obliged to build. My responsibility to abide by the standard compensates for my not having caused the standard and substitutes for my having caused it in the role of making me accountable for the sled’s lack of conformity to it. Even though I do not cause the standard, I have an obligation to act in conformity with it such that when I don’t, the lack of conformity is imputable to me. Implicit in the foregoing suggestion is the idea that something can explain or account for an effect in virtue of not doing what it ought to have done. We often offer such explanations. “Why did she miss the jump shot?” “Because she didn’t ‘square up’ beforehand.” “Why did he fail the test?” “Because he didn’t study.” “Why did the dough not rise?” “Because she forgot to put yeast in.” Such explanations include also the not-doings of nonrational, or nonmoral, agents. “Why did he fall?” “Because the rope didn’t hold.” “Why is the mouse still in the basement?” “Because the cat didn’t catch it.” In such cases, we commonly impute an effect to an agent on account of the agent’s nonperformance. She caused the dough not to rise by forgetting to add yeast. The rope accounted for his fall by not holding. In such cases, it is the fact that the cause, in some sense, “ought” to have performed the act in question that makes its nonperformance explanatory. While we might well explain the mouse’s continued existence in terms of the nonperformance of the cat, we wouldn’t explain it by the nonperformance of the crickets, hopping happily about the basement. Unlike cats, crickets don’t solve rodent problems. Killing mice is not among the things crickets ought to do.38

116

Free Will and God’s Universal Causality: The Dual Sources Account

Recognizing causation by nonperformance conforms, then, to common explanatory practice. It also enjoys philosophical precedent. Consider, for example, the following from Aquinas: One thing proceeds from another in two ways. First, directly; in which sense something proceeds from another inasmuch as this other acts; for instance, heating from heat. Secondly, indirectly; in which sense something proceeds from another through this other not acting; thus the sinking of a ship is set down to the helmsman, from his having ceased to steer. But we must take note that the cause of what follows from want of action is not always the agent as not acting; but only when the agent can and ought to act. For if the helmsman were unable to steer the ship or if the ship’s helm be not entrusted to him, the sinking of the ship would not be set down to him.39

Here Aquinas makes especially clear what we have seen already, namely, that whether an agent ought to have done something is causally or explanatorily relevant. In particular, we can explain an effect by an agent’s nonperformance only when the agent ought to have performed the act in question. Return, then, to the sleigh building examples. In neither scenario do I cause the standards for Olympic bobsleds, and in both scenarios I build a sled that lacks conformity to those standards. But my sled’s lack of conformity is imputable to me only in the second scenario, not the first; for, only in the second scenario is it the case that I ought to have built a sled that so conforms. The fact that I ought to have built a sled that so conforms is what makes it such that my sled’s lack of conformity to the standards can be explained by me, even though I built only the sled and did not also make the standards. Note, however, that there is a slight ambiguity in the presentation thus far. On one way of putting it, I account for the lack of conformity in virtue of my not having built according to the standards, as I ought. On a second way of putting it, I account for the lack of conformity in virtue of my having built a sled that does not conform to the standards by which I was obliged to build. On the first way, which perhaps more neatly fits the examples of causing by nonperformance discussed above, I have a nonperformance—my not building according to the standards—that is explanatorily prior to my sled’s lack of conformity to the standards. The lack of conformity is imputable to me because of what I don’t do. On the second way, there is no explanatorily prior nonperformance. Rather, I account for the lack of conformity in virtue of what I do—my building something that lacks conformity to the standards according to which I ought to have built. In my view, either way of construing the account provides a plausible explanation of why my sled’s lack of conformity is imputable to me, even though I make the sled and not also the standards. As intimated above, this same explanatory framework can be used to show how the sinner accounts for his act’s lack of conformity to the moral standard. The sinner does not cause the moral standard, but he has a responsibility to it. He ought to govern himself in accordance with the moral standard. When he acts in a way that does not conform to the standard, his act’s lack of conformity can,

Does God Cause Sin?

117

therefore, be explained either by his failure to do what he ought (a not-doing) or by his doing something contrary to what he ought (a doing). Either way, his responsibility to abide by the standard compensates for his not having caused the standard and substitutes for his having caused it in the role of making him accountable for his act’s lack of conformity to it. Suppose, for example, that I lie for the sake of avoiding embarrassment, and that the moral standard includes a prohibition against lying. Since I ought to govern myself in accordance with the standard, I ought to have applied the rule against lying by choosing to refrain from telling the lie in question. My failure to govern myself by the moral standard to which I am responsible makes me accountable for the lack of conformity to the standard in virtue of which my lie is sinful.40 Yet, crucially, this same explanatory framework does not imply that God causes the sinful act’s lack of conformity to the moral standard, at least not given an almost universally shared assumption about God. For, given this framework, God would cause a sinful act’s lack of conformity to the moral standard only if God had a responsibility not to cause any creaturely acts that lacked conformity to the moral standard, that is, only if God ought not cause such acts. But if God has such a responsibility, if he ought not cause such acts, then God has manifestly failed to do what he ought. And yet it is an almost universally shared assumption that if God exists, then he cannot fail to do what he ought. It follows from this assumption that if God causes a creaturely act that lacks conformity to the moral standard, then he has not failed to do what he ought; he did not have a responsibility not to cause such acts. By the same token, it follows that if God does not cause the creaturely agent’s applying the moral standard by choosing to refrain from a sinful act, then God’s causing the creaturely agent’s choosing to refrain is not something God ought to have done. But, then, unlike the sinner, our explanatory framework does not imply that God causes the sinful act’s lack of conformity to the moral standard. The lack of conformity is imputable to the sinner, but not to God. There is, then, a principled basis for holding that even though God causes the act of sin and all its positive features, only the sinner, and not God, causes the lack of conformity to the moral standard in which the badness of a sinful act consists. Yet, an important question remains. Granted the uncontroversial assumption that if God exists he never fails to do what he ought, one might think that if an all-powerful, allgood God exists, he ought never to cause creaturely acts of sin, but rather always to cause creaturely choices to refrain from sin. One might then take the existence of moral evil to be evidence against the existence of an all-powerful, all-good God, or at least against the existence of an all-powerful, all good God, who causes all entities distinct from himself (and therefore causes whatever creaturely acts lack conformity to the moral standard). This is the problem of moral evil—a problem left unaddressed by the foregoing account of how God can be said to cause the act of sin without causing sin itself. We turn to this problem in the following chapter. Before doing so, however, one very brief final question about what has gone before. I have argued that God cannot be held to cause the bad act’s lack of conformity to the moral standard simply by causing that act and its positive properties. But, for the very same reason, it would seem that God cannot be held to cause the good act’s conformity to the moral standard, simply by causing the good act and its positive properties. And, yet, it might be thought desirable to maintain an asymmetry that credits God with

118

Free Will and God’s Universal Causality: The Dual Sources Account

causing the good act’s conformity to the standard, even while denying that God causes the bad act’s lack of conformity. Now, it is possible to preserve a kind of asymmetry even if God does not cause the good act’s conformity to the standard. The asymmetry consists in the fact that while both the good act and the act of sin (qua exercise of active power) are caused by God and good, the sinner alone causes the lack of conformity to the moral standard in which the badness of the sinful act consists. Even so, there is a straightforward way to maintain that God does cause the good act’s conformity to the moral standard. For, one can hold that while the lack of conformity to the standard is a privation and, therefore, not an entity, the good act’s conformity to the moral standard—the very thing lacking in the bad act—is an entity, namely, a positively existing relation of conformity to the standard. In that case, however, unlike the privative lack of conformity, God will cause the conformity to the moral standard just as, according to DUC, he causes all entities distinct from himself. He will cause the conformity, not in virtue of causing the good act and its positive properties but, rather directly, in the way he causes all other entities.

7

The Problem of Moral Evil

As discovered in the previous chapter, it does not follow from God’s causing the act of sin that God causes sin itself. Yet, this discovery still leaves open the atheistic charge that the existence of moral evil of the amount and type we find in the world is inconsistent with the existence of a being, like God, who is all-powerful and wholly good. By far the most popular theistic responses to this problem of moral evil incorporate the famous Free Will Defense (FWD), with its central claim that it is not possible for God to create creatures who are free to do good or bad and at the same time ensure that they always refrain from the bad. In this chapter, I argue that given what has been established earlier in the book FWD fails, even if we assume a libertarian conception of creaturely freedom. I then suggest how a theist can respond to the problem of moral evil without FWD. Finally, I argue that the theist who adopts Dual Sources, and recognizes the failure of FWD, is surprisingly at no significant disadvantage in responding to the problem of moral evil when compared to the responses available on the most popular alternative approaches for combining theism and libertarian freedom, namely, Molinism and Open Theism.

7.1 The Failure of the Free Will Defense In the previous chapter, my argument that God does not cause the act of sin’s lack of conformity to the moral standard made use of the uncontroversial assumption that if God exists, he never fails to do what he ought. If God never fails to do what he ought, then it is not the case that for any creaturely act that lacks conformity to the moral standard, God ought not to have caused that act. Nor is it the case that instead of causing the act, God ought to have caused the creaturely agent’s applying the moral standard by choosing against the sinful act. For these reasons, the lack of conformity to the moral standard, though imputable to the sinner in virtue of the sinner’s failing to do what he ought, is not imputable to God in virtue of God’s failing to do what God ought. Yet an obvious question remains: Why doesn’t God act in such a way as to preclude the sin in question? More generally, why does God permit the vast amounts and horrendous types of moral evil we find in the world? And why does God cause acts of sin, even if he does not cause the privations in virtue of which they are sinful? Wouldn’t

120

Free Will and God’s Universal Causality: The Dual Sources Account

an all-good, all-powerful God create a world devoid of moral evil? Doesn’t the existence of such evil count as evidence against the existence of an all-good, all-powerful God? With these questions we confront the problem of moral evil, conceived as an argument against the existence of God based on the seeming incongruity between the world’s moral evil and the existence of a being with the attributes traditionally ascribed to God. One might grant that if God exists, he never fails to do what he ought. But one might also think that a being with the sort of goodness and power traditionally ascribed to God surely ought to have acted so as to prevent the quantity and types of moral evil we find. And so one might reason that such a God does not exist. Of course, theists are not only confronted by a problem of moral evil. Natural evil also abounds. In contemporary literature on the problem of evil, “moral evil” is usually taken to include not only the evil of morally bad actions and failures to act but also the evil that results from these bad acts and failures. If a man commits murder in cold blood, moral evil so defined includes both the evil of the man’s actions and intentions, and the evils of the victim’s suffering prior to death, the victim’s loss of life, the anguish of the victim’s loved ones, the financial deprivations of the victim’s dependents, and so on. “Natural evil,” by contrast, is usually defined to include evil other than morally bad acts and failures to act or the bad results of such acts and failures. Presumptive examples of natural evil so defined include birth defects, children suffering from cancer, hikers being buried by an avalanche, or animals burning in a forest fire, all such examples assuming that the evils are not the result of any morally bad acts or omissions.1 It is beyond the scope of this work to offer a comprehensive response to the problem of evil. But since Dual Sources has significant implications for a theistic response to the problem of moral evil, it is important to say something about these implications and to discuss how they should affect an evaluation of Dual Sources. What, then, are the implications of Dual Sources for a response to the problem of moral evil? In short: if Dual Sources is correct, then the most common theistic response to the problem of moral evil, the so-called Free Will Defense, fails. In the remainder of this section, I will show why Dual Sources implies the failure of FWD. In the following sections, I will consider the bearing of this implication on an evaluation of Dual Sources. Richard Swinburne offers a representative description of FWD: The free-will defence claims that it is a great good that humans have a certain sort of free will which I shall call free and responsible choice, but that, if they do, then necessarily there will be the natural possibility of moral evil. … A God who gives humans such free will necessarily brings about the possibility, and puts outside his own control, whether or not that evil occurs. It is not logically possible—that is, it would be self-contradictory to suppose—that God could give us such free will and yet ensure that we always use it in the right way.2

The point of FWD is to explain how the existence of the sort of moral evil we find in the world is consistent with the existence of an all-powerful, all-good God. The start of the explanation is that God is confronted with a choice between two mutually exclusive alternatives: He can either grant some of his creatures the freedom to perform or refrain

The Problem of Moral Evil

121

from morally bad acts, or he can ensure that no morally bad acts ever occur. He cannot do both. But, continues FWD, the freedom to perform or refrain from morally bad acts is a tremendous good. Thus, it is reasonable to believe that an all-good God might grant some of his creatures this freedom, even though it places outside his control whether these creatures use this freedom for good or bad. With this initial move in place, the proponent of FWD can go on to explain why God permits the bad results of morally bad acts, when he could have prevented these results, and why God allows the choice of what seem to be especially heinous acts, when he might have limited choice to a range of less depraved options.3 FWD, as described here, is generally thought to presuppose an incompatibilist or libertarian conception of freedom, and it is not difficult to see why. If freedom is compatible with determinism, then it seems that God was not confronted with the choice between mutually exclusive options that FWD attributes to him. On the contrary, God could have given us the freedom to perform or refrain from morally bad acts and also ensured that no morally bad acts ever occur. He could have done so by always introducing factors that determined us freely to refrain from evil whenever a choice for evil presented itself. In order to rule out this possibility, most have thought that the proponent of FWD must assume the falsehood of compatibilism and the truth of libertarianism.4 But suppose we assume libertarianism. Does it then follow that God was confronted by a choice between mutually exclusive options, that of granting his creatures the freedom to perform or refrain from morally bad acts, or that of ensuring that no morally bad acts ever occur? I believe by far the prevailing answer to this question is “yes.” But, as I will now show drawing off the conclusions of Chapter 4, this common answer is mistaken. For this reason, it can also be shown that FWD is a failure, at least if satisfying the standard conditions for libertarian freedom is sufficient for genuine freedom. Suppose I tell a lie for the sake of avoiding embarrassment, a morally bad act, I will assume. What would it be to say that God could have “ensured” that this act not occur? I take it that God could have ensured that this act not occur, if he could have done something such that, necessarily, had he done it, I would not have told the lie (i.e., performed this particular act of lying). Now, according to FWD, it is not possible that God grant me the libertarian freedom, in a given set of antecedent conditions, to tell this lie or to refrain from this lie, and also ensure that I freely refrain from telling it. Thus, according to FWD, if God grants me such freedom, he puts outside his control whether I tell the lie in these prior conditions. We can summarize by saying that FWD is committed to the following assumption: FWD Assumption: Necessarily, if, in antecedent conditions c, I was free in the libertarian sense either to perform some sinful act a or to refrain from it, and I freely (in the libertarian sense) performed a, then there is nothing that God could have done such that, necessarily, had he done it, I would have freely (in the libertarian sense) refrained from a in c.

But is this assumption true?

122

Free Will and God’s Universal Causality: The Dual Sources Account

Recall Chapter 4, where the Dual Sources account of divine agency and creaturely freedom was presented. There it was shown that God’s causing some creaturely act A of creaturely agent S is consistent with A’s being free in the libertarian sense. It was also shown that God’s causing A is consistent with S’s having had the ability to do otherwise all antecedent conditions remaining the same and with A’s being ultimately up to S, something for which S is ultimately responsible. Coupling DUC (divine universal causality) and EM (the extrinsic model of divine agency), we inferred that any free creaturely act A and God’s act of causing A are simultaneous necessary conditions for one another. Neither is prior to the other, but it is not possible for one to exist without the other. Since it is not possible for one to exist without the other, A and God’s act of causing A are both logically necessary and logically sufficient for each other. There is no world with A, but not God’s act of causing A, and no world with God’s act of causing A, but not A. Yet, because neither act is prior to the other, God’s act of causing A exists too late to make it such that S had to perform A and could not have done otherwise. This is why God’s causing A leaves S’s ability to do otherwise intact. By the same token, S’s performing A exists too late to make it such that God had to cause A and could not have done otherwise. The simultaneity of their actions means that both God and S retain their abilities to do otherwise all antecedent conditions remaining the same, even though it is not possible for God to cause A without S’s performing A or for S to perform A without God’s causing it.5 In light of these conclusions, we can see that FWD Assumption is false. Suppose that in certain antecedent conditions I was free in the libertarian sense either to tell a lie or to refrain; and suppose I tell the lie. Given Dual Sources, God’s causing my act of telling the lie is consistent with my having had the ability to tell the lie or refrain from it all antecedent conditions remaining the same and also with my having ultimate responsibility for my act of lying. Given Dual Sources it is not possible that I tell the lie and God not cause my act of telling it. But it does not follow that God could not have done something such that, necessarily, had he done it, I would not have told the lie. For, in the actual scenario in which I tell the lie, my telling the lie is not prior to, but rather simultaneous with, God’s act of causing it. Since it is not prior to God’s act of causing it, my act of telling the lie exists too late to make it such that God had to cause my act and could not have done otherwise. Thus, even though there is no world in which I tell the lie and God does not cause it, God could have done otherwise than cause it, all antecedent conditions remaining the same. He could have refrained from causing me to tell the lie. Had he refrained, given that his causing is logically necessary for my telling, my telling the lie would not have occurred. Not only could God have refrained from causing my act of telling the lie. He could also have caused my applying the moral standard in an act of judging that the lie was not to be told. Had he done so, I would have performed the positive act of deciding not to tell the lie, of refraining from telling it. And I would have done so freely, in the libertarian sense. For I would have refrained under conditions in which I was free in the libertarian sense either to tell the lie or to refrain from it, conditions that did not determine me to one or the other. The foregoing shows that, in light of Dual Sources, FWD Assumption is false. Since FWD presupposes this assumption, the foregoing shows that, in light of Dual Sources, FWD fails. Indeed, given Dual Sources, it was possible for God to have given

The Problem of Moral Evil

123

us libertarian free choice to perform or refrain from an indefinitely large number of morally bad acts while ensuring that moral evil never occurred. Of course, it was not possible, consistent with libertarian freedom, that God ensure that moral evil never occur without our in every case freely refraining from morally bad acts, having had the ability to perform them all antecedent conditions remaining the same.6 But the lack of that possibility is consistent with the claim that even in this world where we don’t always refrain from morally bad acts, God could have ensured that no morally bad acts occurred. For, given points discussed in Chapter 4 (Section 4.5) and reiterated above, from the fact that there is no world in which God causes free creatures always freely to refrain from evil without free creatures always freely refraining from it, it does not follow, in a case where free creatures don’t always freely refrain from evil, that God could not have caused their always freely refraining from it. And if God could have caused us always freely to refrain from evil, then FWD fails as a means of exculpating God for the moral evil we find in the world. Contrary to FWD, God did not have to choose between the options of giving us freedom to perform or refrain from bad acts and ensuring a world without moral evil. I have said that FWD fails given Dual Sources. But does it fail simpliciter? I think it does, given two assumptions. The first assumption is that the relations between a free creaturely act A and God’s act of causing A described in Chapter 4, and endorsed by Dual Sources, are possible. For, if they are possible, then FWD Assumption, on which FWD depends, is false.7 And it seems to me that the relations described there are, in fact, possible; I know of no good reason to think otherwise. The second assumption is that the account of human freedom accommodated in Chapter 4 suffices for genuine freedom. That is, it suffices for genuine freedom that an act satisfy the strict, as well as the broad, conditions for being free in the libertarian sense; that it be such that its agent could have done otherwise all antecedent conditions remaining the same; that it be within its agent’s power whether or not the act occurs; and that the act satisfy standard understandings of what is required for its agent’s having ultimate responsibility for the act. If Dual Sources is possible and these characteristics make for genuine freedom, then FWD has not shown that it was impossible for God to give us genuine freedom while ensuring that we always use it for good. Now, some might wish to argue that the foregoing characteristics do not suffice for genuine freedom—that genuine freedom requires that my free acts be up to me and me alone, that they cannot be genuinely free if they are caused by God and are up to him as well as me. But why should we think that? To avoid begging the question, such an argument must point to some feature that is necessary for genuine freedom and that is ruled out if my act is caused by and up to God, as well as me. But what will that feature be? We have already seen in Chapter 4 that my act’s being caused by and up to God is consistent with my having the ability to do otherwise all antecedent conditions remaining the same, with its being within my power whether or not my act occurs and with my having ultimate responsibility for my act. These are the very features taken to be necessary for freedom on the most robust accounts. Moreover, these features surely suffice for moral responsibility, and a free act is often understood simply as an act of the sort for which one can be morally responsible.8 In short, the prospects for arguing

124

Free Will and God’s Universal Causality: The Dual Sources Account

that the kind of freedom accommodated in Chapter 4 is not genuine freedom do not seem very strong. What the foregoing shows, then, is that FWD is, or is very likely, a failure. It isn’t simply that it is ruled out if one adopts DUC. It is a failure because it was possible for God to give us genuine freedom but still ensure that we always use that freedom for good. If God could create us always acting well without undermining our freedom, then “respect for our freedom” cannot be the reason God permits moral evil. Another response to the problem of moral evil will be necessary.9

7.2 Responding to the Problem without FWD If you agree that FWD is a failure simpliciter, then you may wonder how the theist can respond to the problem of moral evil in a way that does not rely on FWD. If you do not agree that FWD is a failure simpliciter, but only that it is ruled out by Dual Sources, then you may wonder how significant a cost it is of Dual Sources that it precludes FWD. These questions can be addressed by considering, first, how a theist might respond to the problem of moral evil in the absence of FWD, and, second, whether this response is significantly less strong than the responses available to the most prominent contemporary alternatives for combining theism and libertarian freedom, specifically, Molinism and Open Theism. I take up the first of these tasks in the present section and the second in Sections 7.3–7.6. One response to the problem of moral evil that does not depend on FWD is to argue that God’s goodness is not, strictly speaking, moral goodness and so does not imply anything about how God will act with respect to creation. As a consequence, the existence of moral evil is consistent with God’s goodness even on the assumption that God could have created a world with free creatures who always do good. Although this approach has been defended by some working within the scholastic tradition,10 I want to pursue the possibility of a response that accommodates the more common view that God’s perfect goodness does include moral goodness with its implication that God acts in ways consistent with moral perfection. Such a response, it would seem, requires identifying certain goods that the existence of moral evil makes possible or goods that presuppose moral evil—goods for the sake of which God may have created a world with the moral evil we find even though he could have created a world in which creatures always act well. In my view, considerable modesty is called for in our attempts to pass judgment on what evils a perfectly good, all-powerful, and omniscient God would or would not permit (and why). My proposal, therefore, is not offered as an account of why God, in fact, created a world with the amount and types of moral evil we find. Rather, I propose a range of goods for the sake of which (or some of which) God might plausibly be thought to have created a world with such evil. I offer, then, what has been referred to as a “defense” not a “theodicy.” And the defense I offer is infused with a healthy dose of skeptical theism. A possible good of the sort needed is suggested by Swinburne’s theodicy for natural evil. According to Swinburne, a good God would create a world in which his free creatures make morally significant choices. Natural evil contributes to this goal,

The Problem of Moral Evil

125

among other ways, by providing occasions for such choices.11 The suffering due to, or threatened by, natural evil makes possible choices between courage or cowardice, compassion or callous indifference, vigilance or neglect. But if natural evil makes possible such choices, so too does moral evil, as Swinburne clearly recognizes when considering and responding to an objection: It may, however, be suggested that adequate opportunity for these great good actions would be provided by the occurrence of moral evil without any need for suffering to be caused by natural processes. You can show courage when threatened by a gunman, as well as when threatened by cancer; and show sympathy to those likely to be killed by gunmen as well as to those likely to die from cancer. But just imagine all the suffering of mind and body caused by disease, earthquake, and accident unpreventable by humans removed at a stroke from our society. No sickness, no bereavement in consequence of the untimely death of the young. Many of us would then have such an easy life that we simply would not have much opportunity to show courage or, indeed, to manifest much in the way of great goodness at all.12

Perhaps surprisingly, Swinburne seems to think that were our suffering limited to that caused by moral evil, there would be insufficient occasion for most of us to exercise morally significant choice. Whether one agrees with Swinburne on this point, it is clear enough that, just like natural evil, moral evil and the effects of morally bad acts provide us with opportunities to make choices of a sort that would not be possible were it not for the suffering caused by these types of evils. Hence, if it is good that we have such opportunities, it is plausible that a good God might create a world including such evils. One might wonder, though, to invert the objection Swinburne considered, why God wouldn’t provide for such opportunities only through natural evil, leaving moral evil out of his creation, given that he could have done so without taking away our freedom. But here it is important to notice that moral evil provides certain challenges and opportunities that natural evil does not provide. Being threatened or wounded by a person acting immorally is not the same as being threatened or wounded by impersonal natural processes. The former always carries with it the additional wound of having been wronged, having been disrespected or hated by another person who ought to have treated one with respect. The virtuous response to being wronged, or to one’s loved one’s being wronged, is more complicated and more difficult than the response to suffering from impersonal causes. To what extent should justice be demanded? What is justice in the particular situation? How can I forgive, even love, the one who has wronged me or my beloved (if that is what virtue requires)? How can I prevent such wounds from leading to a general mistrust of others or even to misanthropy? It is not implausible to think that among the good worlds a good God might have created is a world in which we face morally significant choices and struggles that can only arise as a result of moral wrongdoing. If so, then God may well have permitted moral evil, in whole or in part, to make possible such choices and challenges. Implicit in the foregoing is the idea that there are certain goods that simply won’t be instantiated unless there is moral evil. The good mentioned above is that of human beings’ facing certain morally significant choices and challenges. But there are clearly

126

Free Will and God’s Universal Causality: The Dual Sources Account

other goods as well. Mercy, repentance, forgiveness, atonement, conversion from evil, even just punishment: All of these goods presuppose moral evil and cannot be instantiated without it.13 I do not deny that there are good worlds, which a good God might have created, that lack moral evil and, hence, that lack these goods. But I also don’t think it implausible that a good God might have permitted moral evil in order that such goods be instantiated. Nor, it would seem, does St. Paul: “For God has imprisoned all in disobedience so that he may be merciful to all” (Rom. 11:32).14 Notice, too, that most of these goods come in less or more impressive instantiations. To forgive someone who wrongfully cut in front of me at the grocery store is less impressive than to forgive a drunk driver who killed my friend in an accident. And to do the latter is less impressive than to forgive someone who intentionally killed my friend in cold blood, in order to steal his wallet. The conversion of a horrible sinner is more magnificent than the conversion of one whose sins are only venial. The grandeur of mercy shown is in proportion to the severity of the punishment deserved. It is not only, then, that the instantiation of these goods presupposes moral evil. The degree of goodness of the instantiation is often in proportion to the badness of the moral evil presupposed. I submit that it is not implausible to think that among the good worlds a good God might have created are worlds that include impressive instances of such goods. If so, then God may well have permitted even horrendous moral evils of the sort we find in the world in order that there be great instances of mercy, repentance, forgiveness, atonement, conversion from evil, and perhaps even just punishment. According to traditional Christian theology, Jesus suffered and died so that all sins might be forgiven, and all sinners shown mercy. How wondrous this love when we consider just how depraved are the human beings to whom Christ’s mercy extends.15 I have argued that FWD fails and have suggested how a theist might respond to the problem of moral evil without FWD. In short, there are goods for the sake of which it is plausible to think that a good God may have permitted moral evil of the amount and type we find in the world, even though God could have prevented that moral evil without eliminating human freedom. Some of these goods consist in opportunities for good human action, such as the love of enemies. Some consist in opportunities for good divine action, such as the manifestation of God’s goodness or glory in exercises of divine justice or mercy.16 But does the sort of response available to Dual Sources place the account at a significant disadvantage in responding to the problem of moral evil? In particular, does it place the account at a disadvantage when compared to the responses available to the most popular alternative approaches for combining theism and libertarian freedom? In the following three sections, I argue that it does not by comparing the response I have suggested to the responses that would be available were Molinism or Open Theism true. I assume the reader’s familiarity with both of these alternative approaches.

7.3 Moral Evil, Dual Sources, and Molinism Given Molinism, for every sin that occurs, God knew prior to creation that the creature would sin if placed in those exact circumstances, yet God chose to create the creature and place him in those circumstances anyway. Why? The most natural thing for the

The Problem of Moral Evil

127

Molinist to say is that God places creatures in circumstances where he knows they will freely sin, and permits these sins, in order that certain goods be possible. But, if that is the Molinist’s answer, then the Molinist response to the problem of moral evil is no different than the response I proposed above. A Molinist might argue that a salient difference in the responses available to us is that while Dual Sources holds that it is up to God as well as the free creature what that creature does in any set of antecedent conditions, Molinism denies that God has any control over what the creature would do in those conditions. As a result, a Molinist might argue that God was extraordinarily unlucky in the contents of his middle knowledge. Perhaps, the counterfactuals of creaturely freedom were such that there is no world inhabited by free creatures that God could have created without moral evil as bad, or worse, than the moral evil we find in our world. Perhaps, for every possible free creature, that creature’s counterfactuals of freedom are such that it could not have been created and placed in circumstances where it always refrains from sin. If these suggestions are taken seriously, then it might be argued that the Molinist has an advantage in responding to the problem of moral evil. For, the Molinist could say that the world God created is among the best God could have given the contents of his middle knowledge. But he could say that, on the assumptions of Dual Sources, God could have created a much better world, a world containing much less, or even no, moral evil. To the foregoing line of argument, let me make two responses. First, the suggestion that God’s options are so restricted by the contents of his middle knowledge that he could not have created a world with free creatures that never sin, or that sin far less often and badly than the creatures of our world sin, strikes me as literally incredible. There are an infinite number of possible free creatures and an infinite number of possible circumstances in which they might find themselves. Surely, not every combination of free creatures and circumstances yields a world with moral evil or moral evil of the type and quantity we find in this world. No one—however omnipotent—could be that unlucky! The claim that God permits moral evil for the sake of goods it makes possible is, at least to my mind, a far more plausible claim for the Molinist to make. But, then, we are back to the same response that the Dual Sourcer can give.17 Second, even if we grant (only for the sake of argument) that, given the contents of his middle knowledge, God could not have created free creatures who either never sin or who sin far less often, the Molinist is still committed to this world’s being, on balance, a good world. For, it was open to God not to create at all or to create a world without free creatures. And that’s what a good God would have done, if the only worlds he could have created with free creatures were bad ones. That the Molinist must admit that our world is a good world is significant. For, then, even if we grant what I will call “the incredible hypothesis” about the content of God’s middle knowledge, the Molinist will still be fairly limited in the advantage he can claim over the Dual Sourcer in responding to the problem of moral evil. He will have to concede to the Dual Sourcer that God has created a good world. His only advantage might be in his options for explaining why God didn’t create a supposedly better world, a world with less or no moral evil. Granting the incredible hypothesis, the Molinist could say that God didn’t create a world with less moral evil because God couldn’t,

128

Free Will and God’s Universal Causality: The Dual Sources Account

given the contents of his middle knowledge. The Dual Sourcer can’t avail himself of this explanation. Still, there are other things the Dual Sourcer can say in response to the question why God didn’t create a world with less or no moral evil. He can say that, in light of the goods that moral evil makes possible, a world with less moral evil may not have been a better world. Perhaps, because of the goods moral evil makes possible, a world like ours is actually better than a world with no or less moral evil. Or, perhaps, the value of a world with free creatures who never sin is incommensurate with the value of a world including goods that couldn’t exist without sin, making it impossible to rank one type of world as objectively better than the other.18 Alternatively, the Dual Sourcer could concede, for the sake of argument, that a world with less moral evil would have been better, but deny that, to be perfectly good, God must create a better world if it is possible for him to do so. This denial is plausible, for it is plausible to think that for any world God could create, there is a better world he could create; so the requirement that God create a better world if he can do so is not one that it would be possible for God to fulfill by the creation of any world. And, surely, one might think it is at least as good for God to create a good world, albeit one for which he could have created a better, than not to create at all.19 These latter explanations of why God doesn’t create a world with less or no moral evil are, in my view, more plausible than the explanation that depends on the incredible hypothesis about the content of God’s middle knowledge. Of course, the Molinist can make use of these latter explanations just as much as the Dual Sourcer. The key point, however, is that the Molinist enjoys no real advantage over the Dual Sourcer in responding to the problem of moral evil. The one explanation available to the former but not the latter is less credible than the explanations available to both. The ability to respond to the problem of moral evil, therefore, would be no better were Molinism, and not Dual Sources, true.

7.4 Moral Evil, Dual Sources, and Open Theism Unlike the Molinist God, the God of Open Theism does not know, prior to creating, what any possible free creature would do in any set of circumstances in which it might find itself. Given Open Theism, it is thus false that for any sin of any creature, God chose from the beginning to create that creature and place it in circumstances in which he knew the creature would commit that very sin. Furthermore, unlike on Dual Sources, it is not the case on Open Theism that for any sin freely performed by a creature in antecedent conditions c, God could have done something such that, necessarily, had he done it, the creature would have freely refrained from that sin in those same conditions. These differences might seem to give Open Theism a significant advantage over Dual Sources and Molinism when it comes to answering the problem of moral evil. Indeed, the relatively weaker and more limited conception of God’s providence and sovereignty favored by Open Theists would seem, in general, to make God less responsible for the happenings of the world, less responsible for the good, which might be considered a cost of the account, but also less responsible for the bad, which many would consider a benefit.20

The Problem of Moral Evil

129

On closer examination, however, the advantages that Open Theism can claim in responding to the problem of moral evil are not as significant as they might at first appear.21 For, in order plausibly to respond to the problem, the Open Theist, just like the Molinist and Dual Sourcer, will need to claim that God’s justification in permitting moral evil hinges on the fact that the moral evils make possible certain goods, known or unknown to us, which defeat or compensate for those evils. To appreciate this point, keep in mind that Open Theists have been at pains to emphasize that their conception of God’s knowledge and providence, while less meticulous than the traditional conception, still enables God to exercise a great deal of control over the course of history. As Hasker puts it: God has complete, detailed, and utterly intimate knowledge of the entirety of the past and the present. He also, of course, knows the inward constitution, tendencies, and powers of each entity in the fullest measure. And, finally, he has full knowledge of his own purposes, and of how they may best be carried out. Everything God does is informed by the totality of this knowledge; the guidance he gives, if he chooses to give it, is wisdom pure and unalloyed. Knowing what he knows, God may sometimes know also that the uninterrupted course of natural action and human responses will best serve his deep purposes. He may, on the other hand, know that for this purpose to go forward there is need for his own direct touch and influence, whether recognized or unrecognized, on this or that human personality. Or finally, he may see that for his purpose best to be fulfilled what is called for is his immediate, purposeful intervention in the processes of nature—in other words, a miracle. Whatever God needs to do, he has the power to do; whatever he sees is best to do happens forthwith.22

Let us call these affirmations regarding God’s knowledge and influence “Standard Open Theism.”23 The God described here is one who, based on his exhaustive knowledge of the past and present, typically knows, to a high degree of probability, what at least, broadly speaking, is going to happen in the near future, barring his influence. Moreover, it is a God with the power to exert an influence that would almost certainly change the outcome that would have been likely were he not to exert that influence, so as to secure a desired result. How does this description of God’s knowledge and providential control affect a response to the problem of moral evil? Given the foregoing account, we can safely say that God has known for thousands of years that the tendency of human beings is to sin gravely and often. We can also say that for at least a great many particular sins, God knew for some time beforehand that those sins, or something similar, were very likely to be committed given the conditions leading up to them.24 Moreover, given the affirmations of Standard Open Theism, it seems that God had the wherewithal to prevent a great deal of this moral evil. He could have intervened in particular cases, seeing to it, for example, that a Hitler or a Stalin got buried by an avalanche. Or, acting more globally, God could have exerted an influence that, while not removing the possibility of, or freedom to do, moral evil, would almost certainly have greatly diminished its frequency and gravity. He could have exerted such an influence, for example, by making his presence better known

130

Free Will and God’s Universal Causality: The Dual Sources Account

to human beings (even those who already believe in him) in the way that a vigilant father might make his presence known to misbehaving children; or he could have refashioned human tendencies by grace or some other means, so that we desire what is good more frequently and what is bad less. Given the affirmations of Standard Open Theism, it would seem that God could have done any of these things. Had he done so, the world would, or would almost certainly, contain a great deal less moral evil. So, why hasn’t God done so? Faced with such a question, an Open Theist might be tempted to respond with one of the following claims: Claim A: God could not, consistent with human freedom to do good or bad, have exercised an influence of a sort that would (very likely) have significantly diminished the frequency and gravity of moral evil. Claim B: Even if God could have exercised such an influence consistent with human freedom to do good or bad, he could not have done so without sacrificing some other good, or introducing some evil, so great that it would not be worth the diminishment of moral evil.

It is important to note that the sort of good that would be sacrificed according to Claim B is not a good of the sort discussed in Section 7.2, namely, a good that presupposes moral evil or has moral evil as a necessary condition. If it were that sort of good, then God’s reason for not acting in such a way as to reduce moral evil would basically be the same as that proposed in that section. Claim B is, rather, suggesting that there is some good that does not presuppose moral evil, which would nevertheless be lost were God to act in the ways needed to reduce moral evil. For the sake of readability, I will not reiterate this caveat every time I restate Claim B. Let us call the Open Theist who affirms either of the foregoing claims a More Risky Open Theist, since he envisions God’s creating free creatures as involving a greater ongoing risk of pervasive and appalling moral evil than what Open Theism might seem to require.25 By contrast, let’s call a Less Risky Open Theist an Open Theist who rejects Claim A and Claim B. The Less Risky Open Theist, thus, agrees that God could have, consistent with human freedom to do good or bad, exercised an influence that would have resulted in significantly less moral evil without thereby sacrificing any good, or introducing any evil, so great that it would not be worth seeing moral evil diminished. Now, clearly the Less Risky Open Theist confronts the question, “Why hasn’t God done so?” Since on this less risky version God permits much evil that he could have prevented without costs of the sort imagined in Claim A or Claim B, God needs some justification for this permission of evil. And the Less Risky Open Theist will have an advantage over the Molinist and Dual Sourcer only if he can give a justification not available to these others, that is, only if there is some justifying good that he can point to and the others can’t. What’s more, for the advantage to mean much, the good will need to offer better justification than the justifying goods available to the Molinist and Dual Sourcer. I am aware of no such good. The most likely candidates for justifying goods are ones of the sort introduced in Section 7.2, goods that cannot exist without moral evil

The Problem of Moral Evil

131

and goods the impressiveness of which is in proportion to the severity of the moral evil they presuppose. But, of course, appeal to these sorts of goods is available to all three accounts. A skeptical theistic approach, which denies that we are in a position one way or the other to determine whether and what goods justify God’s permission of sin, is also available to all three. In short, although I cannot absolutely rule it out, it seems unlikely that the Less Risky Open Theist can point to a justification for God’s permission of so much moral evil that is available to his account, but not to Molinism and Dual Sources, much less a justification that is better than the justifications available to Molinism and Dual Sources. Granted, there is this difference. Whereas on Dual Sources (and probably on Molinism) God could have created a world where free creatures always choose the good, the influence that God could exercise on Less Risky Open Theism would not be enough to eliminate moral evil entirely. It would only reduce the frequency and gravity of moral evil. But, once it is admitted by the Less Risky Open Theist that God can be justified in permitting moral evils he could prevent for the sake of the goods they make possible, it is hard to see on what grounds he could object to the Dual Sourcer’s or Molinist’s claim that God, though able to create a sin-free world, permits the moral evils he does for the goods they make possible. Let us now consider More Risky Open Theism. I’ll make three broad points. First, Claim A and Claim B strike me as more doubtful than plausible. Regarding Claim A, the denial that God can exercise, in a way consistent with human freedom to choose between good and evil, the kind of influence I suggested above would significantly reduce moral evil is highly questionable. I can think of only two reasons that might support the claim. One is that God’s influencing things by making his presence better known to us would interfere with our autonomy or independence, in a way that undermined significant freedom.26 A second is that God’s refashioning human tendencies or desires would undermine significant freedom on the grounds that, as Swinburne says, “in order to have a choice between good and evil, agents need already a certain depravity, in the sense of a system of desires for what they correctly believe to be evil.”27 Given this necessary “depravity,” it might be argued that if God refashioned our inclinations so that we desired the good more frequently and the bad less, then he would undermine the conditions for significant choice. Yet, neither reason is convincing. God’s making us more aware of his presence would not take away our ability to commit sin, just as a father’s making his presence known to misbehaving children does not take away their ability to continue misbehaving. It would just make sin less likely.28 And even if one needs some desire for actions that are evil in order to choose them, such choices would still be possible if human beings desired bad actions with less intensity than we do and good actions with more—a change in our desires that would almost certainly result in considerably less moral evil in the world. Moreover, the claim that significant freedom rules out the sort of divine influence in question seems inconsistent with the traditional Christian understanding of humanity’s prelapsarian condition. According to Christian tradition, Adam and Eve had significant free choice, even though God made his presence much more known to them than he does to most of us, and even though, prior to the fall, their desire for the bad was far less than is most of ours now. If we take the Christian story as a guide, it, thus, does not appear that significant freedom

132

Free Will and God’s Universal Causality: The Dual Sources Account

would be undermined by God’s making his presence more known to us or by his doing more to refashion our inclinations and desires.29 Regarding Claim B, it is far from obvious what good would be sacrificed or evil introduced were God to exercise the sort of influence suggested, such that seeing moral evil diminished would not be worth the cost.30 Certainly, in discussions of the problem of divine hiddenness, theists have proposed reasons why God might not make himself better known, or his presence more felt, in ways that I have suggested would lead to a reduction in moral evil. The proposals are many and controversial, and are often so conjectural in their speculations about the possible effects of God’s making his presence better known that reaching a secure judgment about the matter would likely prove difficult even after protracted discussion.31 Would, for example, God’s making his presence better known so as to reduce moral evil result in people’s acting better only for the wrong reasons? Would it have a deleterious spiritual effect by dampening our desire for God or increasing spiritual complacency? While such results do not seem to me especially likely, I do not know how to rule them out. But, even if the exact costs and benefits of God’s making his presence better known remain in doubt, there is still the question of why God couldn’t exercise an influence that would diminish sin by means of refashioning human tendencies so that we desire what is good more frequently and what is bad less? What good would be sacrificed in this case? Would such refashioning undermine the regular order of nature? But it certainly seems that God could exert this influence without rendering nature chaotic or unpredictable. Christians believe that God does something like this for some people through grace, and it is not generally thought that God’s doing so undermines the regularity of nature. Is God maxing out the grace by which he can heal disordered human desires? Is he maxing out the grace by which he can do so consistent with natural regularity? Maybe, but I don’t see an especially good reason to think so.32 Indeed, it is common for believers to pray for grace, for themselves and others, seemingly on the assumption that God could give more than he is giving. This exploration of possible reasons in support of Claim B has been brief, but it is perhaps enough to explain my initial skepticism. The second broad point I wish to make concerning More Risky Open Theism holds independently of doubts about Claim A and Claim B. The point is that by further weakening the influence that God can (consistent with human freedom and the preservation of other goods) exercise over history, it would make the Open Theist’s claim to be offering a biblical conception of God’s providence and influence even less plausible than its critics already think it is.33 It would also adversely affect the credibility of the Open Theist’s assurances that, on their view, God has the wherewithal to achieve his ultimate goal for creation, a loving community of friendship among human beings and between human beings and himself. Hasker offers such assurances by noting how improbable it would be for everyone to reject God’s offer of friendship.34 But, presumably, the Open Theist takes God’s goal not to be that just one or a few will accept God’s offer but that most or all will, at least in the end. If that is God’s goal, the prospects of God’s being able to accomplish it seem pretty dim, given the limited influence affirmed by More Risky Open Theism. By contrast, if that is God’s goal, the prospects look at least somewhat better given the influence possible on Less Risky Open Theism, and the prospects look excellent on Molinism or Dual Sources.

The Problem of Moral Evil

133

The third and final broad point is that, absent appeal to compensating goods that so much sin makes possible, More Risky Open Theism would hardly help the Open Theist in responding to the problem of moral evil. For, absent such an appeal, it would make God’s initial creation of significantly free human beings an act with too great a chance of ending in disaster, an act of ill-advised risk unbefitting a wise and good creator.35 And, having had ample time to observe how significantly free human beings behave under the conditions of his limited influence, it would arguably make God’s continuing to bring forth and sustain them an act even more foolhardy, unless moral evil is compensated for by goods it makes possible.36 In addition to its other problems, then, adopting More Risky Open Theism is not going to help the Open Theist address the problem of moral evil, at least not unless it is combined with appeal to compensating goods made possible by the evil. But compensating goods are the sort of thing that the Molinist and Dual Sourcer can invoke to justify God’s permission of sin and which the Less Risky Open Theist, it seems, must also invoke. This similarity brings us to the crucial question of what advantage, if any, the More Risky Open Theist can claim when it comes to justifying God’s permission of moral evil. Abstracting from (but not forgetting!) the problems raised in the first two broad points above, I think that More Risky Open Theism does, in fact, have an advantage over the other positions in justifying God’s permission of moral evil, though how much of an advantage is a matter of debate. I have argued that, absent appeal to compensating goods made possible by moral evil, on More Risky Open Theism God would likely not be justified in originally creating and continuing to bring forth significantly free human beings, since, given the terms of More Risky Open Theism, God’s doing so would involve an ill-advised risk, unbefitting a good and wise Creator. But I’ve suggested that God might well be justified if there were compensating goods that moral evil made possible; for, then, even if, under the conditions of God’s limited influence, significantly free human beings sin badly and often, the results would not be disastrous, since the sins would be offset by the goods that presuppose them. Why, then, does God permit so much sin according to More Risky Open Theism? Here the More Risky Open Theist can give an answer not available to the Molinist and Dual Sourcer (or to the Less Risky Open Theist, for that matter). He can say that God is justified in permitting so much sin both because it is the price of creating significantly free human beings (or, alternatively, of preserving some other good, or avoiding some other evil, as on Claim B) and because the many and horrible sins predictably committed are offset by compensating goods. This answer is not available to the Molinist and Dual Sourcer since, on their accounts, God could, without sacrificing any good that does not presuppose moral evil, create significantly free human beings and not permit any moral evil at all or anything like the amount and type we find in this world.37 How significant is the More Risky Open Theist’s advantage in responding to the problem of moral evil? That depends on further judgments. One will think the advantage very significant if one judges that the goods that sin makes possible are not by themselves sufficient to justify God’s permitting so much sin, but that these goods would be enough to justify it together with either

134

Free Will and God’s Universal Causality: The Dual Sources Account

a. the continuing existence of significantly free human beings who (as God has known for a long time)38 tend to use their freedom badly; or b. the preservation of some good that does not presuppose moral evil, which good would have to be sacrificed were God to exercise an influence that would diminish moral evil; or the avoidance of some evil that would be introduced were God to exercise such an influence. On the other hand, if one judges that the compensating goods are sufficient to justify God’s permission, then one will think the More Risky Open Theist has at most a slight advantage. For, given this latter judgment, one will agree that the God of Molinism and Dual Sources is justified in terms of the compensating goods. The slight advantage enjoyed by the More Risky Open Theist would come from the fact that, on his account, God has an additional justification for permitting sin, in that his not acting so as to prevent more of it is a requirement either of leaving human beings significantly free or of preserving some good that would have to be sacrificed (or avoiding some evil that would be introduced) were God to act so as to reduce the amount and gravity of sin. I can imagine different judgments. But, in my view, the More Risky Open Theist has only a small advantage. For it seems to me that the compensating goods are plausibly thought enough by themselves to justify God’s permission. And, if we had good reason to think they were not by themselves sufficient, it is hard to judge whether either the value of significantly free human beings most of whom sin with such frequency and gravity or the value of some good that God would have to sacrifice to prevent moral evil would be enough to make up for the justificatory deficit. But whatever one’s judgment about the advantages of More Risky Open Theism in responding to the argument from moral evil, we have seen in the first and second broad points above that there are significant problems with the account, even when compared to Less Risky Open Theism. In my view, these problems weigh heavily against it.

7.5 Sin and the Divine Will Thus far, I have argued that Less Risky Open Theism enjoys no advantage over Molinism and Dual Sources in responding to the problem of moral evil and that More Risky Open Theism has only a minimal advantage, even abstracting from its other problems. But it might be argued that I am overlooking an important advantage that can be claimed by Open Theists of either sort. For even if one agrees that Molinism and Dual Sources are roughly on par with Open Theism when it comes to justifying God’s permission of sin, it might be thought that Open Theism is better able to accommodate the idea that sin is something contrary to God’s will, something of which God disapproves. Hasker has suggested that if Molinists hold that sin is included within God’s providential plan for the sake of the goods it makes possible, then they are not entitled to claim that God merely “permits” sin. According to Hasker, to include sin for the sake of these goods would be to will or intend it as a means to an end and to will or intend is not merely to permit. Hasker allows that Molinists could save the language of “permission,” if they denied that God includes sin for the sake of goods it makes

The Problem of Moral Evil

135

possible. But, then, the Molinist would lack a justification for God’s permission of these sins. So, the Molinist faces a dilemma: either God wills and intends sin as a means, rather than merely permitting it, or God lacks justification for his permission of sin.39 Similarly, Hasker argues that on a no-risk view of providence, “every … crime and atrocity is exactly what God desires for it to be,”40 not that each of these crimes is what God would prefer if considered in isolation but that each is something God desires when considered in the broadest possible context of creation as a whole. Thus, no-risk views imply that God approves of and desires the occurrence of sin, when surely it would be better to say that sin runs contrary to God’s will and desire. If these objections hold against Molinism, they likely hold also against Dual Sources. On Dual Sources, for any sin committed in a given set of antecedent conditions, God could have done something such that necessarily, had he done it, the sinner would have freely refrained from the sin in those very same conditions. If God chose as he did for the sake of a good the sin made possible, then it looks like he willed the sin for the sake of that good and that he desires the sin’s occurrence. On the other hand, if God’s choice was not for the sake of some compensating good, then God lacks justification for permitting the sin. Notice that the extent to which the Open Theist can avoid this dilemma depends on the type of Open Theism adopted. Consider, first, Less Risky Open Theism. It is true that, on this view, specific instances of sin are not always included within God’s plan, as they presumably are on Molinism (at least once we set aside the incredible hypothesis); nor are they permitted when God could have done something such that, necessarily, had God done it, the sinner would have freely refrained, as on Dual Sources. Still, God could have exercised an influence such that, consistent with significant human freedom, the frequency and severity of sin would almost certainly have been drastically reduced. So, why doesn’t God do so? Here, it looks like the Less Risky Open Theist faces a dilemma very similar to the one Hasker advances against the Molinist. Either God permits much sin that he could have prevented, but without justifying reason, or he permits it for the sake of goods that such sin makes possible. And, if the latter, then the God of Less Risky Open Theism is as vulnerable (or not) to the charge that he wills and desires sin, as is the God of Molinism or Dual Sources. Perhaps, the Less Risky Open Theist could deny that God wills specifically this particular act of murder or this particular sacrilege. But he couldn’t, while accusing the Molinist’s God of willing sin, fairly deny that God wills that there be plenty of murder and sacrilege, and the like. For the God of Less Risky Open Theism could prevent much murder and sacrilege, but, if he is justified, refrains from doing so because of the goods these sins make possible. Thus, if the Molinist’s God wills specific instances of murder and sacrilege as a means to goods, the God of Less Risky Open Theism wills that there be plenty of murder and sacrilege as a means to the goods for the sake of which he chooses not to act in ways that would reduce the occurrence of these sins. The More Risky Open Theist has an easier time evading the dilemma. Although he must acknowledge that God’s justification for permitting so much moral evil hinges on the compensating goods it makes possible, he can say that the reason God permits so much moral evil is not the goods it makes possible but that God’s acting to prevent more of it would either take away significant human freedom or require sacrificing

136

Free Will and God’s Universal Causality: The Dual Sources Account

some other important good. Accordingly, he can claim that God would prefer a world with no sin at all, but that God is willing to create and continue a world in which he knows there will be frequent and horrendous sin because the sins are compensated for by goods they make possible, with the consequence that such a world is, overall, still a good place. On this account, the compensating goods might be said at least partially to justify God’s permission of so much moral evil, yet without its being the case that God permits the occurrence of moral evil for the sake of these goods. And if God does not permit the occurrence of moral evil for the sake of these goods, then there are no grounds for the charge that God wills moral evil as a means or that God intends or approves of moral evil.41 Is the ability of More Risky Open Theism to evade the dilemma a point in its favor when compared to Dual Sources and the other approaches? That depends on the extent to which it is genuinely problematic to hold that God permits moral evil for the sake of goods it makes possible. One reason for thinking it not problematic is that, on multiple occasions and in important contexts, scripture presents God as permitting (or something stronger) moral evil for the sake of good. We have already cited St. Paul’s teaching at Rom. 11:32 that God has consigned all to disobedience so that he might be merciful to all.42 Another example is Gen. 50:20, where Joseph, hearing the apology of his brothers who had sold him into slavery, remarks, “Even though you intended to do harm to me, God intended it for good.”43 Another example is Acts 4:27–28, where we read: “For in this city, in fact, both Herod and Pontius Pilate, with the Gentiles and the peoples of Israel, gathered together against your holy servant Jesus, whom you anointed, to do whatever your hand and your plan had predestined to take place.”44 Those who take scripture as a measure for what we can say regarding God’s relationship to moral evil should proceed with caution before ruling out accounts on which God permits moral evil for the sake of good.45 But does God’s permitting moral evil for the sake of good imply that God wills moral evil, as Hasker suggests? Interestingly, thinkers such as Augustine, Anselm, and Aquinas understood permitting to be a type of willing.46 This makes sense given that permissions are motivated. When one permits something, one has a reason for doing so, and to act or to refrain for a reason engages the will. At the same time, and though he clearly thinks that God permits moral evil for the sake of goods,47 Aquinas also denies that God wills moral evil.48 Should we say, then, that God wills moral evil, or not? Well, certainly not without significant qualification. What God wills are certain goods that presuppose moral evil. But he does not will or approve of the moral evil in its own right. If I will that my daughter develop the virtue of courage, I might permit her to experience fear or a difficult situation, which I could easily have prevented, in order that she have an opportunity to develop that virtue. If one insists that I will the fear or the difficult situation, then it will be necessary to acknowledge a merely permissive form of willing, whereby the object willed is neither approved of nor produced by the person willing it, but merely allowed by that person, when he could have prevented it, in order that some other good of which he does approve be possible. In similar fashion, God’s “permissive willing” of sin is consistent with other senses in which God does not will sin, and even senses in which sin is contrary to God’s will and something against which God issues commands.

The Problem of Moral Evil

137

Before I expand on this last point, let me say a bit more about what is involved in God’s permitting sin according to Dual Sources, which task will help us see how God’s permissive willing relates to other sorts of divine willing. As preparation, consider that Dual Sources enables us to distinguish between divine action that is antecedent to, concurrent (or simultaneous) with, and consequent to a given human action. A divine action is antecedent to a given human action if it is the bringing about of something that is temporally or explanatorily prior to the human action. A divine action is consequent to a given human action if it is the bringing about of something temporally or explanatorily subsequent to that human action. A divine action is concurrent or simultaneous with a given human action if, as understood on EM and explained in detail in Chapter 4, the divine action simply consists in the bringing about of the human action, such that both actions are logically necessary and sufficient for each other, but neither is prior to the other. As we saw in Chapter 4, where divine and human action is concurrent or simultaneous, the human act can be caused by God and still satisfy all the conditions for being free in the libertarian sense, including being such that its human agent could do otherwise all antecedent conditions remaining the same and such that its human agent is (together with God) ultimately responsible for the action. This concept of concurrent divine action, made possible by EM, is central to the Dual Sourcer’s reconciliation of libertarian freedom and traditional theism.49 Recall that, on EM, God’s causal acts are also acts of willing; God can be said to will his effects since he brings them about on purpose, for a reason. But, then, since God’s causal acts can be antecedent, concurrent, or consequent relative to a given human action, the same is true for the divine acts of willing that are identical to these acts. We know from the previous chapter that a “sin of action” consists in two elements: an act (the “act of sin”) together with that act’s privative lack of conformity to the moral standard. God does not cause “sins of action,” since he does not cause the privations in virtue of which the sins of action are sinful and in which their sinfulness consists. God permits sins of action. But, given DUC, it is not possible for God to permit a sin of action without causing the “act of sin,” that is, without causing the really existing exercise of active power that, qua exercise, is good. Indeed, on Dual Sources, God’s permissive willing of a sin of action just is God’s causing the act of sin for whatever reason he permits the sin, whose vitiating element (the privation) God does not cause. Notice that since God’s causing the act of sin is concurrent or simultaneous with the act, so also is God’s permissive willing concurrent with the sin of action. Dual Sources, therefore, rejects the notion of an antecedent permissive decree, understood as a choice or decision of God’s to permit a sin that is explanatorily prior to the sinner’s sinning (and which decree also “determines” the sinner to sin in cases where the decree is understood to be not only explanatorily prior but also logically sufficient for the sin). Nevertheless, since Dual Sources acknowledges that God could have done otherwise than cause the act of sin, and that had he done otherwise, the sin of action would not have occurred, the account is committed to God’s permitting the sin for a reason. And, again, this permissive willing of the sin is nothing distinct from God’s causing of the act of sin for whatever reason God permits the sin. Often (maybe always) when we commit a sin of action we do so contrary to some direction that God gives us in a divine act that is antecedent to our sin of action. For

138

Free Will and God’s Universal Causality: The Dual Sources Account

example, I might tell a lie in the face of an antecedent divine act of issuing a command against lying, or of creating me with a nature such that lying is contrary to my good, or of giving me a grace to avoid lying.50 Indeed, God’s permitting someone’s disobedience, for whatever reason he may have, presupposes an antecedent divine act in light of which what that person does can be characterized as disobedience. Suppose, for example, we take St. Paul’s suggestion that God permits disobedience in order that he might show mercy (Rom. 11:32). And suppose the disobedience in question is my telling a lie. In such a case, God’s causing my act of lying is concurrent with my act and constitutes God’s permissive willing of my sin of disobedience in order that God show mercy to me. God’s showing mercy by forgiving my sin would then be consequent to my telling the lie. But for my lying to constitute disobedience in the first place there must also be a divine act, such as a command, antecedent to my telling the lie, in light of which my telling it constitutes disobedience. Thus, we can distinguish an antecedent divine willing or act of direction against my telling lies, expressed, for example, in a divine command; a concurrent divine permissive willing of my lying and disobeying God in order that God might show mercy; and a consequent divine willing of mercy to me in God’s forgiving of my disobedience. If it is possible for God permissively to will a sin of disobedience, then it must be possible for God permissively to will that we sin against the direction given in some antecedent divine act. Despite strong precedents within the tradition, the foregoing sort of analysis may well seem alien to some. But is the analysis incoherent? Against it, one might object that it is illegitimate to hold that God merely “permits” sin, when God causes the act of sin. One might contend that, on this analysis, God intends the occurrence of sin, in violation of the moral prohibition against intending evil. One might, furthermore, doubt the consistency of God’s commanding an agent not to do what God permissively wills that the agent do. Finally, one might think that this analysis precludes our saying that sin is contrary to God’s will and judge this preclusion to reduce the analysis to absurdity.51 It seems to me, however, that the analysis can withstand these objections. Why say that God “permits” sin even though he causes the act of sin? “Permits” is appropriate for two reasons. First, and primarily, “permits” suggests that although (a) God does not cause the sin, nevertheless (b) he could have acted in such a way as to prevent it, both of which points hold true on Dual Sources. Second, “permits” or “permissively wills” suggests, better than any alternative of which I am aware, that God does not approve of sin in its own right, even though he lets it be on purpose when he could have prevented it. Does God, on Dual Sources, intend the occurrence of sin in virtue of the fact that he permits it on purpose for the sake of the goods it makes possible? Well, certainly, God doesn’t intend the occurrence of sin as an end. God’s end is the good for the sake of which he permits the sin. Nor is the Dual Sourcer necessarily committed to the claim that God intends sin as a means, at least not if a “means” is understood to be something an agent uses to bring about or advance its end. If, for example, God permits instances of sin in order that there be instances of forgiveness, the forgiveness is not brought about through the instrumentality of the sin, as an inflated tire is brought about through pumping air. Rather, the sin is only a necessary condition of instances of

The Problem of Moral Evil

139

forgiveness, such that any world God creates that includes instances of forgiveness will be a world that includes moral evil. If God creates such a world, then he permits the moral evil for the sake of the forgiveness (i.e., the forgiveness is his reason for allowing the moral evil). But that doesn’t make the moral evil a means (or instrumental good) used for bringing about God’s end. Whether scripture interprets God as intending the occurrence of sin as a means (understood in the sense above) is, perhaps, unclear. But even if it does, it is doubtful that God’s doing so would flout the moral norm against intending evil as a means to good. That norm cannot be interpreted as a prohibition against willing the existence of an evil state of affairs as a means. Were that the case, it would be prohibited to will that someone’s legs be separated from the rest of their body in order to save their life. Rather, the norm against intending evil as a means to good is properly understood as a prohibition against intentionally or deliberately doing evil for the sake of good, that is, against doing what is in itself morally wrong in pursuit of a good end. But, of course, the Dual Sourcer (with support from scripture and tradition) can simply deny that for God permissively to will the occurrence of moral evil is for God to do evil. How can we make sense of God’s commanding an agent not to do what God permissively wills that the agent do? The charge of incoherence here depends on taking, for example, God’s command that we not lie and God’s permissively willing our lies, as entailing that God both wills that we lie and wills that we not lie, in the same respect or in some absolute, unspecified sense. But we can deny this entailment. For God to command that we not lie is not for God to will that we not lie in an absolute or unspecified sense (nor, for that matter, is God’s permissive willing of our lies willing lies in an absolute or unspecified sense). Rather, the command directs us to what we ought to do, either by placing us under an obligation to do it or by indicating what it is good for us to do as conducive to the flourishing of the nature God has given us. Now, there is nothing incoherent about God’s directing us to what we ought to do and then permissively willing our acting contrary to what we ought to do for the sake of some good that our sin makes possible. It certainly does not follow that if God permissively wills our acting contrary to what we ought to do, then we actually ought to have acted so contrary or that God really thinks that we ought to have acted so contrary! Just because something is good for God to permit does not mean that it is good for us to do or that it is what we ought to have done! Perhaps, it will be maintained that God’s commanding that we not lie communicates God’s absolute will that we not lie, such that God’s permissively willing the sin of lying contradicts God’s command or makes of God a deceiver who implies he wants one thing when he really wants another. But I would simply deny that God’s commands communicate such an absolute, unqualified will. They communicate God’s will regarding what we ought to do, but not God’s will that in every case we succeed in doing what we ought. God’s command, therefore, is not incompatible with God’s permissively willing our failing in what we ought nor does it involve any deception on God’s part if he permissively wills our failing. The foregoing suggests, also, a response to the last of these objections, namely, that Dual Sources undermines our ability to affirm that sin is contrary to God’s will. Because the Dual Sourcer allows that God permissively wills sin, he indeed cannot say that sin is contrary to God’s will in an absolute, unspecified sense. Nevertheless,

140

Free Will and God’s Universal Causality: The Dual Sources Account

there are meaningful senses in which the Dual Sourcer can say that sin is contrary to God’s will. For example, he can say that sin is contrary to God’s will in the sense that God disapproves of sin in its own right, despite permitting it. Contrast this with other things, such as acts of justice, which God approves of in their own right.52 Moreover, there is a clear sense in which to disobey God’s command is to act contrary to God’s will. For the command expresses God’s will concerning what we ought to do. Thus, in acting contrary to God’s command, we act contrary to God’s will regarding what we ought to do: We either fail to do that which God wills we ought or we do that which God wills we ought not. But doing that which God wills we ought not can, without contradiction, be permissively willed by God for the sake of some good purpose. There is one sense in which what we are doing is contrary to God’s will—we are doing what God wills we ought not. There is another sense in which it isn’t—God permissively wills our doing what we ought not. Both senses are true and meaningful, but, because the senses are distinct, there is no contradiction.

7.6 God’s Involvement in Sin: A Cost of Dual Sources? I have been arguing that Dual Sources is not at a significant (if any) disadvantage, compared to Molinism and Open Theism, when it comes to responding to the problem of moral evil. But it might be thought that I have neglected a glaring disadvantage: the degree of God’s involvement in sin. After all, as explained in Chapter 6 and as acknowledged just above, even though God does not cause sin on Dual Sources, it is an implication of DUC that God causes all acts of sin. If a mobster breaks the legs of an uncooperative business “partner,” the Dual Sourcer (and any proponent of DUC) must say that God causes the mobster’s choice and whatever the mobster does to break them or whatever entities are involved in the mobster’s doing so. And, of course, human beings perform many acts that are even worse than unjustly breaking someone’s legs. I take it that all the accounts we have been discussing agree that God is involved in sin at least to the extent of upholding and sustaining the various natural powers by which sinners perform their sinful acts. After that, it seems that the accounts can be ordered, from most to least, in terms of God’s degree of involvement. The highest degree of involvement is found on Dual Sources, not only because, for any sin, God could have acted such that the sinner would have freely done otherwise under those same conditions but also because, as we’ve been reminded, God causes the very act of sin (even if not the sin itself). The next highest degree is found in Molinism, on which God places human beings in circumstances where he knows they will commit the very sins they do.53 Then there is Less Risky Open Theism, on which God would appear to be considerably less involved in sin than on Dual Sources and Molinism, but where it is still the case that God refrains from acting so as to reduce the amount and gravity of sin for the sake of goods that the world’s sins make possible. The least involvement is found on More Risky Open Theism, where God does not permit sin for the sake of particular goods they make possible, but where God does ultimately approve of and permit a world mired in sin because, though he would prefer that it be sinless, the sins

The Problem of Moral Evil

141

are compensated for by goods they make possible, such that the world is still a good place overall.54 Putting aside whatever else might be said for or against the various accounts, it seems that, when it comes to responding to the problem of moral evil, an account on which God has a higher degree of involvement in sin will be at a disadvantage compared to an account on which God has a lower degree of involvement. But just how serious a cost for Dual Sources is God’s relatively high degree of involvement in sin? One might think the cost not especially serious, at least if we assume that our understanding of God’s goodness must be consistent with the level of God’s involvement in sin found in scripture, for scripture not infrequently presents God as highly involved in sin. Not only do passages such as Gen. 50:20 and Acts 4:27–28, discussed in the previous section, depict sins as falling very much within God’s plan and purpose. In these and other passages, God’s role vis-a-vis the occurrence of sin seems to be far more active than simply refraining from preventing what it was within God’s power to prevent. Thus, the prophet Isaiah asks, “Why, O Lord, do you make us stray from your ways and harden our heart, so that we do not fear you” (Isa. 63:17). In Exodus God is depicted as hardening the heart of Pharaoh so that he will not let the Israelites go in defiance of God’s very command (Ex. 4: 21–23). St. Paul reaffirms this episode, explicitly rejecting that it and the like are incompatible with God’s justice (Rom. 9: 11–21). In Amos we read: “Does disaster befall a city, unless the Lord has done it” (Amos 3:6)? In Isaiah we read: “I am the Lord, and there is no other. I form light and create darkness, I make weal and create woe; I the Lord do all these things” (Isa. 45:6–7). In Lamentations we read: “Who can command and have it done, if the Lord has not ordained it? Is it not from the mouth of the Most High that good and bad come” (Lam. 3:37–38)? While the dreadful happenings God is said to bring forth in these last three passages may or may not refer to sins per se, they would certainly seem to involve creaturely agents acting sinfully. Thus, Old Testament scholar John Hayes, commenting on the passage from Amos, notes that Amos was contending that “the present trouble being undergone by Israel, its oppression by neighboring states, was the work of Yahweh, a work already begun and which would culminate in the future devastation of the northern kingdom.”55 Thus, John Piper remarks that “the calamities in view [in these three passages] involve human hostilities and cruelties that God would disapprove of even as he will that they be.”56 Admittedly, passages such as those cited likely underdetermine any particular philosophical-theological account. Nevertheless, it may be considered an advantage of Dual Sources that while denying that God causes sin, it allows that God causes the act of sin and that God permits sin on purpose, thus accommodating (however roughly) a level of divine activity in relation to sin suggested by such passages.57 The degree of God’s involvement in sin suggested by these passages fits rather better with Dual Sources and (perhaps) Molinism than it does with More Risky, or even Less Risky, Open Theism. At the very least, one who takes scripture as a measure ought to pause before insisting that Dual Sources gives us a picture of God’s involvement in sin that is incompatible with God’s goodness. That point seems right even if one finds it perplexing how God’s involvement in sin is compatible with God’s goodness.

142

Free Will and God’s Universal Causality: The Dual Sources Account

Even apart from scriptural considerations, it is doubtful that God’s degree of involvement in sin on Dual Sources puts the position at a significant disadvantage relative to the alternatives. On Dual Sources, but not on the other accounts, God causes the act of sin; for example, God causes whatever it is the mobster does to break his victim’s legs. And we might understandably wonder how causing such acts could be consistent with moral perfection. Yet, on all the accounts God sustains the powers employed by the mobster when he breaks the legs. And we might also wonder how that could be consistent with moral perfection. Surely, we don’t think it would be consistent with moral perfection for a human being, say, to empower the mobster by handing over the baseball bat to be wielded in the leg breaking. Proponents of all the views will argue that God’s sustaining the powers by which the mobster breaks his victim’s legs is justified because it is a necessary condition of his giving us significant freedom and responsibility. But an analogous argument can be made by the Dual Sourcer with respect to God’s causing the act of sin. Just as God is justified in sustaining the powers of the mobster, a necessary condition of the good that is the mobster’s significant freedom and responsibility, so God is justified in causing acts of sin, which, given DUC, is a necessary condition of the goods for the sake of which God permits sins of action. This analogous argument is one against which it would be difficult for proponents of the alternative views to lodge a credible protest. As we’ve seen, in responding to the problem of moral evil, Molinists and Less Risky Open Theists, just like Dual Sourcers, need to say that God permits sins that he could have prevented without loss of human freedom for the sake of goods those sins make possible. And even More Risky Open Theists must hold that God continues a world mired in sin largely because the sins are compensated for by such goods. Proponents of the alternative views might try to argue that it is consistent with God’s moral perfection that he permit sins in light of the goods they make possible, but nevertheless not consistent with God’s moral perfection that he cause acts of sin. Yet what could justify this claim, especially in a case where God’s causing the acts of sin is a necessary condition of permitting the sins of which these acts form a part? No doubt, moral perfection rules out a human being causing an act of sin, his own act, or (were it possible) someone else’s—such as the mobster’s act of breaking someone’s legs. But all the accounts are going to have to recognize a difference between what is consistent with human goodness and what is consistent with God’s goodness. It is not permissible for a human being to supply the mobster with a bat for breaking legs. If such a person justified himself by saying he was simply empowering the mobster so that the mobster had significant freedom and responsibility in the matter (which he would not have were a weapon not available to him), most would rightly think it a lame excuse. And, yet, this same justification is commonly and not implausibly used for God’s sustaining the powers by which the mobster commits his crime. Nor is it consistent with human goodness, say, to permit assaults that one could easily prevent for the sake of goods they make possible; yet, on Molinism and Less Risky Open Theism, as on Dual Sources, God permits assaults for the sake of such goods— at least he does so if he is justified in permitting them. And on More Risky Open Theism, God continues a world where such crimes are rampant largely because they are compensated for by such goods.

The Problem of Moral Evil

143

All the accounts need to recognize a difference between what is consistent with the goodness of human beings and what is consistent with the goodness of God. But once this need is admitted, the fact that it is inconsistent with human goodness to cause acts of sin is no longer a strong reason for thinking that God’s causing such acts is inconsistent with God’s goodness. And I can’t think of another plausible reason, especially once it is granted that God is justified in permitting sins of action for the sake of goods they make possible and once it is seen that, given DUC, God’s permitting such sins is going to involve his causing the acts which, together with the privations, constitute these sins. The distinction between what is consistent with human goodness and what is consistent with God’s goodness raises large questions, which I do not have space to pursue. One question concerns the reason for the distinction. It presumably derives from other differences between God and us: for example, the difference between God’s authority and our authority, the difference between the scope of God’s responsibility (the whole of creation) and the scope of our responsibility (not that), the difference between the extent of God’s knowledge and the extent of our own, and the difference between what the Creator owes to human creatures and what we owe to each other. Another question concerns what the distinction implies about the extent to which we can speak meaningfully about God’s goodness, given that to call God perfectly good and to call a human being perfectly good will have diverse implications for the types of actions they can perform. These are interesting and important questions, but not ones that need to be answered for present purposes.58 My purpose in this section has been to argue that God’s high degree of involvement in sin on Dual Sources does not place the position at a significant disadvantage, when compared to the alternatives, in addressing the problem of moral evil. My first argument is that scripture presents God as highly involved in sin, in a way that fits rather well with Dual Sources. My second argument is that the God of Dual Sources is justified in causing acts of sin because he is justified in permitting sins for the sake of goods they make possible, and, given DUC, his so permitting requires his causing the acts that in part make up those sins. Proponents of the alternative approaches are in no position to object to this argument. Molinism and Less Risky Open Theism are likewise committed to God’s being justified in permitting sins for the sake of good they make possible. And all the positions must recognize a distinction between what actions are consistent with God’s goodness and what actions are consistent with human goodness—a recognition that militates against the charge that God’s causing the acts of sin needed to permit sins for the sake of goods they make possible conflicts with God’s goodness. Further discussion of the distinction must await another time. Before closing the section, however, let me briefly apply some of the foregoing considerations to address whether, on Dual Sources, God’s relation to sin conflicts with God’s justice and love. To begin, one might think it unjust or unfair to the sinner if God’s causing the act of sin, and permissively willing the sin, determined or necessitated the sinner’s sinning, rendering the sinner unable to avoid the sin and its consequences. But, on Dual Sources, because God’s causing the act of sin and permissively willing the sin are concurrent or simultaneous with the sinner’s sinning, it

144

Free Will and God’s Universal Causality: The Dual Sources Account

is not even possible for these divine acts to occur unless the sinner sins when the sinner could have done otherwise all antecedent conditions remaining the same. The sinner, then, is as ultimately responsible for his sin and its consequences as on any libertarian account. Not only is the sinner’s sinning ultimately up to the sinner, God would not seem to owe it to a creature defectible by nature, to prevent it from acting defectively or to refrain from concurring with its defective action. The charge of injustice to the sinner, then, is not compelling. More pressing might be whether Dual Sources makes God unjust to the victims of sin. Yet Dual Sources does not present a unique problem in this regard; the problem must be faced by any account on which God permits sins he could have prevented for the sake of goods they make possible or even by accounts on which God only sustains the powers employed by sinners while committing their sins. As we have seen, the theist will need to allow that God’s unique relationship to creatures, and to creation as a whole, renders consistent with God’s justice a level of involvement in sin that most would deem unjust if exercised by one creature in relation to another creature. Nor would it seem incompatible with God’s loving a creature that God permissively wills that creature’s sin; nor inconsistent with loving the victims of sin that God permits them to be victims. Scripture presents God as loving those whose disobedience he permits in order that he might return mercy upon those same sinners. And it is a common theme within Christianity, with definite scriptural roots, that God is hard on those he loves, permitting them to endure much affliction and suffering. I have contended that God’s relation to sin on Dual Sources is consistent with how that relation is portrayed in scripture. Suppose that contention is correct. Then those who think it suffices for compatibility with God’s justice and love, that an account of God’s relation to sin is consistent with how that relation is portrayed in scripture, ought judge the Dual Sources account of God’s relation to sin to be compatible with God’s justice and love. In this chapter, I have argued that, given Dual Sources, FWD fails. Yet, the failure of FWD does not preclude a plausible response to the problem of moral evil. Not only can the Dual Sourcer give a response, his response is arguably as (or nearly as) plausible as those available on the most popular alternative approaches for combining theism and libertarian freedom, Molinism and Open Theism. Perhaps, the most significant concern about Dual Sources will be its understanding of God’s relationship to moral evil. This and the previous chapter have attempted to face that concern square on, to be honest about the implications of DUC and Dual Sources, and to show that the concern is not fatal for the overall view. In Chapter 8, we will return, in the context of a discussion of predestination, to the question of whether Dual Sources is compatible with God’s justice and love.

8

Providence, Grace, and Predestination

In defending the extrinsic model of divine agency (EM) in Chapter 5, I appealed to claims that might appear to make trouble for an account of God’s knowledge. The present chapter begins by considering the problem and arguing that EM actually solves it by simultaneously constituting an extrinsic model of divine agential knowing, a model whereby God knows all his creaturely effects immediately in the course of knowing what he is doing. With this model in hand, Dual Sources offers an interesting variation on the Eternity Solution to the Freedom-Foreknowledge Dilemma: It is because all points in time are present to God’s eternity that God can have knowledge of future free acts where the agent could do otherwise all antecedent conditions remaining the same; but, unlike on the conventional Eternity Solution, God’s knowledge of free creaturely acts has dual sources—not just the creature but also God—and is ultimately up to both. The chapter continues by showing how Dual Sources provides an account of meticulous divine providence, consistent with libertarian freedom, yet without the doctrine of middle knowledge. It shows how Dual Sources makes possible libertarian friendly, yet robust, accounts of grace and predestination on which not just the possibility of meritorious action but the meritorious action itself—and ultimately our salvation—is God’s gift. It shows that, contrary to a possible worry, Dual Sources affords a viable understanding of divine–human dialogue. And it shows how Dual Sources can accommodate a variety of positions on the question of Heaven and Hell. The chapter, in sum, exhibits the power of Dual Sources to offer new solutions for reconciling libertarian freedom with the claims of traditional theology.

8.1 An Extrinsic Model of Divine Knowing As we have seen, the key to reconciling DUC with libertarian creaturely freedom is EM, the extrinsic model of divine agency. In Chapters 4 and 5, I argued that EM is not ad hoc. Rather, it is an integral part of scholastic theology since EM is the model of divine agency best able to accommodate the doctrine of divine simplicity and the doctrine that God is not really related to creatures. Among the tenets of divine simplicity is the denial that God has intrinsic accidents or accidental properties. In Section 5.2, I offered independent support for this denial in the course of arguing that, even apart from divine simplicity, DUC together with plausible assumptions entails EM.

146

Free Will and God’s Universal Causality: The Dual Sources Account

Yet the claim that God lacks intrinsic accidents may be thought problematic when it comes to God’s knowledge. For, assuming that God is omniscient and that there are contingent truths, it follows that God has contingent knowledge: (CK): God knows contingent truths.

Yet one might assume that God’s knowing some truth T implies something intrinsic to God, such as God’s belief that T or perhaps just God’s state or act of knowing T. What’s more, one might suppose that such states would not exist were God not knowing T. Thus, we have the following thesis about God’s knowledge (the knowledge thesis): (KT): Necessarily, God’s knowing some truth T implies some entity intrinsic to God that would not exist were God not knowing T.

Let us take as an example of a contingent truth God knows: 1. Donald Trump is the 45th president of the United States. Necessarily, God knows (1) only if (1) is true. Since (1) is a contingent truth, it follows that God’s knowing (1) is contingent. But, by (KT), God’s knowing (1) implies some entity intrinsic to God that would not exist were God not knowing (1). Since there are possible worlds in which God does not know (1), there are worlds in which this entity does not exist. So, God’s knowing (1) implies some contingent entity intrinsic to God. But this entity cannot be identical to the divine substance, for God is a necessary being: (NB): The divine substance exists necessarily.

The only other option is for the entity to exist as an intrinsic accident or accidental property of God. In short, (CK), (KT), and (NB) entail that God does have intrinsic accidents, contrary to divine simplicity and to what I argued in the course of defending EM.1 Fortunately, the problem can be dissolved by rejecting (KT) and embracing an extrinsic model of divine knowing, a model according to which God’s contingent knowledge, which varies across worlds, does not involve any intrinsic variation in God. On such a model, whatever variations are required across worlds by God’s varying contingent knowledge are outside of, or extrinsic to, God. There are various extrinsic models of divine knowing.2 On an extrinsic belief model, God’s knowledge is a species of belief, and all of God’s beliefs are occurrent, true, and meet whatever other conditions are required for beliefs to constitute knowledge. But, on this view, God’s beliefs are not intrinsic to God. Rather, God’s beliefs consist in relations to propositions that exist outside of God, and these relations do not involve any intrinsic states of God that would be otherwise were the relations not holding, were God not so believing. It follows that were God not knowing (1), God would be intrinsically no different.

Providence, Grace, and Predestination

147

Some may be puzzled by the belief model’s contention that believing could be a mere relation to a proposition. Yet, philosophers have sometimes identified mental acts or states with relations to various objects. Prominent accounts of perception, for example, view perceiving as a relation.3 And some have argued, similarly, that believing is a relation. Michael Thau, for instance, observes that “the philosophical conception of belief according to which beliefs … are internal states of subjects is usually assumed to be an utter triviality with nary a thought that it might be a substantive claim for which argument is needed.”4 But, says Thau, “once you see that beliefs are propositions and believing is nothing more than a relation to a proposition, there’s no reason to think that the absence of internal belief states would make the existence of beliefs in any way less robust.”5 A proponent of the belief model can take comfort in the fact that his move is not unprecedented, even apart from theological concerns. Still, the belief model is not my favored extrinsic model of divine knowing. For one thing, it seems to presuppose a Platonism about propositions that is arguably theologically problematic.6 For another, it appears to fall short of the most perfect type of knowing. William Alston, for example, has argued that the highest type of knowledge, the one we should attribute to God, is a type on which the knower is in direct, unmediated contact with the object known: Immediate awareness of facts is the highest form of knowledge just because it is a direct and foolproof way of mirroring the reality to be known. There is no potentially distorting medium in the way, no possibly unreliable witnesses, no fallible signs or indications. The fact known is “bodily” present in the knowledge. The state of knowledge is constituted by the presence of the fact known. … Since God is absolutely perfect, cognitively as well as otherwise, His knowledge will be of this most perfect form.7

If one grants this ideal, an obvious problem arises for understanding God’s knowledge as a species of belief. Beliefs take propositions as their objects, but the concrete universe is not made of propositions. Propositions may represent or correspond to the concrete universe, but they are not the thing itself. In order to secure the highest form of knowledge for God, we need an account on which God’s knowing is not a species of believing and on which his knowledge of concrete reality is direct, not mediated by propositions or anything else. The immediate cognition model is an attempt to supply such an account. In its basic structure, it is very much like the belief model. The difference is that instead of having acts of believing consisting in relations to contingently true propositions, God has acts of knowing consisting in relations to the contingent realities known. Like the belief model, these relations do not involve any intrinsic states of God that would be otherwise were the relations not holding, were God not so knowing. Thus, the model is consistent with the denial that God has intrinsic accidents. It is easy to see why the immediate cognition model would be attractive to someone who endorses the sort of cognitive ideal introduced above. Just as on the belief model a divine act of believing essentially consists in the proposition believed (along with the believing relation to it), so on the immediate cognition model God’s act of knowing

148

Free Will and God’s Universal Causality: The Dual Sources Account

essentially consists in the contingent object known together with the knowing relation. God’s knowledge of contingent reality is unmediated in the strongest sense possible. Rover—all four paws and 63 pounds—directly constitutes (with the knowing relation) God’s state and act of knowing him. God’s cognitive state, his act of knowing, extends out beyond God to embrace the contingent things in themselves, and those contingent realities, in turn, directly inform God’s acts of knowing.8 The immediate cognition model is an improvement over the belief model, but a question remains for them both. While both tell us what God’s contingent acts of knowing consist in, they do not explain how God is in cognitive contact with the contingent objects he knows. Presumably, he is not in cognitive contact by being acted upon by contingent objects, for being acted upon likely involves being affected intrinsically and being affected intrinsically would involve introducing or removing contingent accidents in God. Perhaps, God’s knowing contingent objects is just a brute fact, with no more to be said about how God is in contact with them. Still, it would be to its advantage if an extrinsic model could tell us more about God’s cognitive contact. Fortunately, Dual Sources has the resources from which to construct just such a model, a model, moreover, on which God’s knowledge of contingent objects is unmediated, just as it is on the immediate cognition model. The key to this model— the agency model—is the idea that God’s activity is inherently cognitional; God directly and immediately knows what he is doing in the doing of it, what he is bringing about in the act of bringing it about. As Barry Miller puts it, “[God] knows Socrates in the very act of creating him.”9 As McCann states, “God’s knowing the universe and his creating it constitute one and the same act.”10 If God knows contingent objects in the act of intentionally bringing them about and if, as EM maintains, God’s intentionally bringing an entity about does not involve anything intrinsic to God that would not exist were God not bringing that entity about, then God can know contingent objects without there being any intrinsic accidents in God. Yet, unlike the belief and immediate cognition models, the agency model constitutes an extrinsic model on which the question of God’s cognitive contact has a clear answer: God is in cognitive contact with all contingent objects in virtue of intentionally and purposefully bringing them about. What is more, God’s knowledge of contingent objects is direct and unmediated. For, God’s act of knowing a contingent object is the same as his act of bringing that object about. But, given EM, his act of bringing it about consists essentially in that object together with its relation of dependence on God. Accordingly, for God to know immediately what he is doing is for God to know the contingent object immediately, and this object directly informs God’s acts of knowing every bit as much as it does on the immediate cognition model. A key assumption of the agency model is that intentional activity is inherently cognitional. When I act intentionally, I know what I am doing in the very doing of it. I don’t need a separate act of knowing to know what I am doing.11 Are there objections to this assumption? One might object that I know only what I am intending to do, not what I actually do, on grounds that what I actually do may fall short of, or go beyond, my intention. This objection would allow that I know what I am intending in the intending of it, but not necessarily what I am doing in the doing of it, even if my doing is intentional.

Providence, Grace, and Predestination

149

Fortunately, even if one accepts this distinction, it does not threaten the agency model. Given EM, God does not intend one thing and do another. As we saw in Chapter 4, his intending to bring about a creaturely effect E just is his bringing about E for a reason: His bringing about contingent objects is his intending them and his intending them is his bringing them about. Hence, if an agent can know what he is intending in the intending of it, God can know in the bringing about of contingent objects, these objects that he is intending by that very same act. Another objection might agree that intentional action (or at least intending) is inherently cognitional, but deny that this phenomenon makes sense without some intrinsic state of the agent that serves as the means by which the agent knows what he is doing. On the standard agent-causal account, there is a plausible candidate for this intrinsic state in the intention, the bringing about of which is the agent’s basic action.12 Since, on EM, God lacks such a state, it may be thought impossible for him to know what he is doing simply in the doing of it. Yet, in response, it can be pointed out that this objection really just begs the question by assuming that cognitive states must be intrinsic states or that all cognizers cognize by means of intrinsic states. This assumption is precisely the proposition that the extrinsic models of divine knowing deny, and I know of no good argument compelling us to accept the assumption. In the case of human actions, it might be thought necessary to introduce some intrinsic state to explain cases where our cognition does not correspond to reality. But, whether or not stipulating such states is necessary to make sense of the possibility of human error, the rationale for stipulating such an intrinsic state is completely absent in the case of a cognizer like God, whose cognition never errs. Similarly, intrinsic states of intention might make sense where one’s action can depart from one’s intention. But, on EM, as we have seen, God’s action and intention do not come apart. Other objections to the agency model are ones that have already been addressed in previous chapters. For example, it might be worried that if God knows our actions in the act of causing them, then none of our actions is free, or that God becomes a cause of sin, or that the problem of moral evil becomes unanswerable. Previous chapters have addressed these concerns. But is the agency model consistent with what has gone before? One might wonder, in light of our recourse to the privation account of moral evil in Chapter 6, how God knows privations. Privations are contingent, but in Chapter 6 I insisted that God does not cause the privations in virtue of which sinful acts are sinful. So, it looks like not all contingent objects of knowledge can be known by God in God’s act of causing them. The solution, I think, is to maintain that when it comes to privations, God does not know them directly, in acts of causing them, but rather indirectly, by knowing, first, what contingent entities ought to exist relative to a given subject in specified circumstances, and, second, what contingent entities actually exist, which God knows in acts of bringing them about.13 For example, if I tell a lie, God knows the lack of conformity to the moral standard in which the sinfulness of my act consists by knowing the act of sin (which he knows in bringing it about) and the moral standard to which this act lacks conformity and to which all human acts should conform.14 Consider one final objection, which will serve as our transition to the next section. We think God knows (or knows about) Julius Caesar and his activities, but we wouldn’t

150

Free Will and God’s Universal Causality: The Dual Sources Account

normally say God is causing them. If something doesn’t exist, then God isn’t causing it. But if God isn’t causing Caesar and his activities, and so on for other past and future substances and events, then the agency model won’t work as an account of God’s knowledge of the past and future. To answer this objection, we need to say something about God’s relation to time.

8.2 Time, Foreknowledge, and a Variation on the Eternity Solution Interestingly, an objection very similar to the foregoing has been leveled by William Hasker against Alston’s proposal, discussed above, that God’s knowledge is “constituted by the presence of the fact known.” Hasker agrees that “immediate, intuitive knowledge, in some form very similar to that portrayed by Alston, does indeed represent a cognitive ideal.”15 The problem, says Hasker, is: Temporal entities may … be immediately, “bodily” present in God’s awareness. But they can only be so present at the times when they exist to be present! But of course, God’s knowledge of such realities cannot be so limited. It follows, then, that there is a requirement for an inner mental representation on God’s part, to enable him to know what has passed away or (perhaps) what is yet to come.16

Hasker’s critique of Alston clearly applies to both the immediate cognition and the agency models. On the former, God’s contingent acts of knowing consist in relations to the contingent objects known, and these objects directly inform and partially constitute God’s acts of knowing them. On the latter, God knows contingent objects in the same acts by which he brings them about, and these acts consist in causal relations to those objects. But, if only the present exists, then only the present is there for God to be related to, and so neither model will accommodate God’s knowledge of the past and future.17 One response to this presentist objection agrees that only the present exists, but wonders whether it follows that God can’t stand in the knowing or causal relation to past and future events. Does the proponent of this objection wish to endorse a proposition such as the following? (NR): Nothing can be related to a past or future object or event.

Yet, this proposition would likely cause problems for the presentist, too. One might suppose, for example, that the “inner mental representation” Hasker thinks necessary for God to know the past or future must stand in a matching or, at least, a representing relation to the past or future event it represents. Similarly, the proposition “Caesar crossed the Rubicon” is presumably true because of a correspondence, made-trueby, or supervenes-on relation to the past event which it is about. In short, even if only the present exists, it is not obvious that an account of knowledge of—or of true propositions about—the past and future can do without some sort of relations to past and future events. Taking such concerns seriously, a presentist might be tempted to

Providence, Grace, and Predestination

151

reject (NR). To do so, however, is to remove the barrier to a presentist’s accepting the immediate cognition or agency model. At the very least, once relations to the past and future are admitted, an argument is required why God can’t stand in knowing or causal relations to past and future events.18 A second response concedes that if only the present exists, then the immediate cognition and agency models won’t work for God’s knowledge of the past and future. It denies, however, that only the present exists—at least for God. In one form the second response rejects presentism in favor of an eternalist account of time on which all moments of time exist, absolutely speaking. In another form it maintains that the question “Which times exist?” does not admit of an absolute answer; rather, the answer is relative to a reference frame or perspective. Relative to our temporally bound reference frame, only the present exists. Relative to God’s eternal reference frame, all moments of time exist.19 Both forms of the response, then, reject the premise that the past and future do not exist for God to be related to. Dual Sources does not offer a new account of time or of God’s relation to time; nor does it offer a new defense of a currently available account. A proponent of Dual Sources must, rather, borrow and incorporate one of these available accounts on which all moments of time are present to God, accounts that are given defenses elsewhere by their proponents.20 Dual Sources affirms that God is eternal, where that is taken to mean (i) that God never began nor will he ever cease to exist, (ii) that God is intrinsically immutable, such that there is no succession in God, and (iii) that all points in time, if any time exists, are present to God. Because all points in time are present to God, God can stand in causal relations to entities no matter their temporal location. Since God knows such entities in his acts of bringing them about, it follows that God can know them whether they exist in the past, present, or future. Moreover, because God is intrinsically immutable, lacking internal succession, and because his acts that bear on creatures are present to his eternity regardless of their temporal location, God’s life is not stretched out over time, with some portions of it having passed him by and others yet to come. Rather, all that he ever does is always present to his eternity. Appeal to God’s eternity is, of course, one of the traditional ways of responding to the so-called Foreknowledge Dilemma, a dilemma that attempts to establish an incompatibility between God’s having infallible knowledge of our future actions and our actions being free. It may seem that if God infallibly knows (or has infallible beliefs about) what his creatures will do in the future, then, for any action we will perform in the future, God has known that we will perform it. But, so the dilemma goes, we can’t now do anything to affect the past and, thus, can’t now do anything to affect what God has known that we will do. Moreover, it is not possible that God infallibly know (or believe) that we will do such and such, and yet we not do that very thing. Since there is nothing we can now do to affect what God has known about what we will do, and since that knowledge is incompatible with our doing something else, it looks as if we can never do otherwise than the things God knows we will do. But if we can never do otherwise, then neither are we free.21 The Eternity Solution halts the dilemma before it gets started. An eternal God’s knowledge (or belief) is not past relative to what we will do at some later time. Thus,

152

Free Will and God’s Universal Causality: The Dual Sources Account

even if it is true that no one can affect the past, it doesn’t follow from the fact that God’s knowing that we will do such and such is incompatible with our not doing it, that we could not do otherwise. God’s eternal knowledge of what we will do is comparable to the knowledge one person has of what another is doing in the present. While it is true that Elizabeth’s knowing what Cecilia is presently doing is incompatible with Cecilia’s not doing it, it does not follow that Cecilia could not be doing otherwise. It only follows that were Cecilia doing otherwise, Elizabeth’s knowledge would not be the same. Linda Zagzebski doubts whether appeal to God’s eternity really solves the Foreknowledge Dilemma on grounds that we have no more control over the eternal realm than we do over the past.22 After all, the eternal realm, like the past, would appear to be fixed and what is fixed would appear beyond our power to affect. In response to this fixity of eternity objection, defenders of the Eternity Solution have argued that we do in fact exercise control over God’s eternal knowledge. We do not do so by changing God’s knowledge, making him know something different than what he knew at an earlier point. Eternity is not susceptible to change. But, say these defenders, we do exercise control over God’s knowledge since what we do is explanatorily prior to God’s knowledge of it. It is because I drank coffee this morning that God knows from eternity that I drank coffee. Had I drunk orange juice instead, God would have known from all eternity that I drank orange juice. God’s eternal knowledge may be fixed, but it is fixed by what we do. What we do explains why God eternally knows what he does regarding our action. And we have counterfactual power over God’s eternal knowledge. Had we done otherwise, God would, from all eternity, have known otherwise.23 Can Dual Sources and the agency model avail itself of this response to the fixity of eternity objection? Not without modification. Recall from Section 4.4 that on Dual Sources, neither a free creaturely act A nor God’s act of causing A is prior to the other; rather, they are simultaneous (or concurrent) necessary conditions of each other. God’s act is not prior to A, for God’s act just consists in A and A’s relation of dependence on God. A is not prior to God’s act, since God must already be causing A in order for A to exist. But, if A is not prior to God’s act of causing A, then it cannot explain or account for that act. And this makes the foregoing response to the fixity of eternity objection unavailable to the agency model. For, on the agency model, God’s act of knowing A is identical to God’s act of bringing A about. If, then, A cannot explain God’s act of bringing A about, neither can it explain God’s knowledge of A. Dual Sources and the agency model cannot, then, simply adopt without revision the foregoing response to the fixity of eternity objection. But the Dual Sourcer can adopt a variation on the response. For even though my free acts do not explain God’s knowledge of those acts, it is false that I have no say or control over God’s knowledge of them or that I am in no way responsible for God’s knowledge. Take any one of my free acts and God’s act of causing it. As explained in Chapter 4, God’s act of causing my act simply consists in my act with its relation of dependence on God. It follows that without my concurrent co-operation in performing my act, God’s act of causing my act does not occur. What’s more, none of the conditions on which my act depends, and not even God’s causing my act, makes it such that I had to perform my act and could not do otherwise. For any one of my free acts, then, I could have done otherwise all antecedent conditions remaining the same, and had I done otherwise, God’s act of causing my act

Providence, Grace, and Predestination

153

would not have occurred. I, thus, have counterfactual power over God’s acts of causing my free acts. But, on the agency model, God’s acts of causing my acts and God’s acts of knowing them are one and the same. So I also have counterfactual power over God’s knowledge of my free acts. I can do otherwise all antecedent conditions remaining the same and were I to do otherwise, God would know otherwise. On Dual Sources, God’s knowledge of my free acts is ultimately up to me, just as it is on the conventional Eternity Solution. But the conventional Eternity Solution operates according to the prevailing assumption that if God’s knowledge of my free acts is ultimately up to me, then it cannot also be ultimately up to God. The assumption is false. Given Dual Sources, God’s knowledge of my free acts is ultimately up to both God and me just as my free acts were shown to be in Chapter 4. God’s knowledge of my free acts is up to God, since God knows my free acts in his acts of bringing them about, which are up to God. Yet God’s knowledge of my free acts is also up to me, since I have counterfactual power over God’s bringing about of my free acts: God brings about (and thus knows) my free acts only with my concurrent co-operation, when I could have done otherwise all antecedent conditions remaining the same. We could say, then, that God’s knowledge of our free acts has dual sources, both God and us. Similarly false is the common assumption that either God’s knowledge of my acts causes those acts or my acts cause God’s knowledge of them. The problem is that a cause must be prior to its effect, but, given Dual Sources, neither my acts nor God’s knowledge of them is prior to the other. To be more precise, what is denied here is that God’s knowledge of my acts as actually existing (as opposed to being merely possible) could be an efficient cause of those acts or that my acts could be an efficient cause of God’s knowledge of them. These denials still leave open the possibility that God’s knowledge of my acts considered as possibilia (in a moment logically prior to God’s causing them) serves as an exemplary cause of my acts. Indeed, the Dual Sourcer is committed to interpreting the traditional claim “that God’s knowledge is the cause of things” in terms of exemplary rather than efficient causality. For, given Dual Sources and the agency model, God’s knowledge of his effects as actually existing is not prior to his effects and so cannot efficiently cause those effects. But God’s knowledge of his effects as possibilia presumably is prior to his effects, and thus this knowledge could serve as their exemplary cause.24 How in the end, then, does Dual Sources account for God’s knowledge of what free creatures will freely do? The answer is threefold. First, God knows free creaturely actions in his acts of bringing them about, as described by the agency model. Second, all points in time are present to God’s eternity, and thus God is present to his bringing about of free creaturely actions, in which he knows them, no matter what point in time those actions occur. Third, our counterfactual power over God’s causing and knowing our free actions explains why it doesn’t follow from (a) God knows what I will do five minutes from now, and (b) it is incompatible with God’s knowing what I will do five minutes from now that I not do that very thing, that (c) I cannot do (have done) otherwise than what God knows I will do five minutes from now. Granted, it is not possible, on the supposition that God knows I will do x, that I not do x. Still, it is up to me whether God knows I will do x since I could do otherwise all antecedent conditions remaining the same, and were I to do otherwise, the supposition that God knows I will

154

Free Will and God’s Universal Causality: The Dual Sources Account

do x would have been false. Of course, as we have seen, on Dual Sources it is also up to God whether God knows I will do x, since God knows I will do x in the act of bringing my doing x about when God could do otherwise all antecedent conditions remaining the same. God’s knowledge of what I will freely do has two sources—it is ultimately up to me and God.25 Notice, finally, that the claim that we have counterfactual power over God’s acts of causing and knowing our free actions poses no threat to the doctrine of divine aseity nor to the doctrine that God is pure act. The doctrine of divine aseity denies that God exists in dependence on other things. The doctrine that God is pure act denies that there is any potentiality in God for ceasing to exist or for existing differently. But, on the extrinsic models of divine agency and knowing, God’s acts of causing and knowing our free acts are not God nor are they even intrinsic features of God. Thus, even though it is not possible, say, that God’s act of causing and knowing my act A exists without my performing A when I could have done otherwise all antecedent conditions remaining the same, it does not follow that it is not possible for God to exist, or for any intrinsic feature of God to exist, without my performing A when I could have done otherwise. My counterfactual power over certain of God’s acts does not make God himself exist in dependence on me nor does it suggest any potentiality in God to cease to exist or to exist differently than he does. Even if God himself does not depend on me, does it violate divine aseity to hold that God’s acts of causing and knowing my free acts depend on me in the way described by Dual Sources? I don’t think so. In the first place, consider that the fundamental rationale for divine aseity stems from the recognition of God as cause of all being apart from himself: Since a cause must be prior to its effect, but a thing can’t be prior to that on which it depends for its existence, it follows that the cause of all being apart from himself cannot exist in dependence on anything. Nothing about this rationale for divine aseity suggests any problem with my having counterfactual power over God’s acts of causing and knowing my free acts. These acts of God, unlike God himself, are not the cause of all being apart from themselves. Indeed, as we’ve seen, God’s act of causing a creaturely act A is not even the cause of A. Our counterfactual power, then, does nothing to flout the logic behind divine aseity, which precludes a cause’s depending for its existence on the very thing it causes. In the second place, the kind of dependence on which God’s acts of causing and knowing our free acts have on Dual Sources is not unlike the sort of dependence they will have on any account. Keep in mind that, on Dual Sources, while God’s act of causing and knowing my act A does not cause A, neither does A cause God’s act of causing and knowing it. God’s act, then, does not depend causally or explanatorily on my act. Rather, it depends only strictly on my act, meaning that there is no world in which God’s causing or knowing my act exists without my act existing. But no account can deny this sort of strict dependence; no account can hold that there are worlds in which God causes and knows my act, but my act does not exist. The sort of dependence here is innocent. Of course, most contemporary libertarian theists seem to be willing both to deny that God causes our free acts and to hold that God’s knowing our free acts is causally or explanatorily dependent on them. What Dual Sources introduces in this context is the idea (i) that, courtesy of EM, God can cause our free acts and

Providence, Grace, and Predestination

155

yet we still have the ability to do otherwise all antecedent conditions remaining the same and (ii) that, courtesy of the agency model of divine knowing, God knows our free acts in his acts of causing them. Add (i) and (ii) to the widespread and innocent recognition that God’s causing and knowing a thing strictly depend on that thing and it follows that we have counterfactual power over God’s causing and knowing our free acts: We could have done otherwise all antecedent conditions remaining the same and had we done otherwise, God’s acts of causing and knowing our acts would not have occurred.26

8.3 Providence A doctrine of divine providence concerns God’s control or direction over the course of history, the extent to which the events of history are in God’s hands and reflect God’s purposes. We gave evidence in Section 2.1 that scripture presents God’s authorship of history to be all-encompassing. Even those, such as Open Theists, who interpret scripture as favoring a less encompassing account of divine providence typically acknowledge that on the traditional view of providence “God’s plan is … all-inclusive … this includes human actions and decisions too … nothing happens that God’s will has not ordained”;27 “every event plays its role in [God’s] grand design.”28 Such a robust, traditional account of providence is often thought inconsistent with libertarian creaturely freedom. For it is often assumed that if our free acts are ultimately up to us as libertarians claim, then they cannot be ultimately up to God in the way that robust providence seems to require. On this assumption, libertarian freedom limits God’s providence, marking out an enormous range of events—all free creaturely acts— to which God’s providence does not extend. God does not incorporate our free acts within his plan for each one of us and for the world; rather, he watches and reacts. His providence consists in trying to influence independent actors to accord with his plan; and when entirely apart from his purposes they fail to do so, his providence consists in trying to influence them yet again or concocting a new plan to achieve his goals. As Open Theist Clark Pinnock puts it: “God has the power and ability to be … an ‘ad hoc’ God, one who responds and adapts to surprises and to the unexpected. God sets goals for creation and redemption and realizes them ad hoc in history. If Plan A fails, God is ready with Plan B.”29 Open Theists, like Pinnock, find this more limited account of divine providence attractive. Not only do they think it helps in responding to the problem of evil, as discussed in Chapter 7. They also view the God of traditional providence a kind of micro-manager or control freak; in their view, a more magnanimous God would relinquish some of his influence. Thus, according to Pinnock: God is a superior power who does not cling to his right to dominate and control … but willingly surrenders power and makes possible a partnership with the creature. … God gives us a role in shaping what the future will be. He is flexible and does not insist on doing things his way. God will adjust his own plans because he is sensitive to what humans think and do.30

156

Free Will and God’s Universal Causality: The Dual Sources Account

At points Pinnock’s description, to my mind, seems better to capture the virtues of a collaborative provost than it fits the God of traditional theism. Certainly, for proponents of DUC, there can be no question of God’s relinquishing his control or influence, not because God is a control freak (as one’s provost might be) but because it is not possible for anything other than God to exist unless it is immediately caused by him. But even if, putting DUC aside, we asked which picture of God’s providence is more attractive, it is not clear why one would prefer the Open Theist’s portrait. For, as we saw in the previous chapter, the Open Theist’s claim to have an advantage in responding to the problem of moral evil is exaggerated. Moreover, on the Open Theist’s account of providence, the very things most important in human life—our free interactions with each other and with God—are ultimately outside God’s hands. Isn’t it a more hopeful picture to have these interactions under the providence of a good and wise God, however opaque to us his purposes may sometimes be? For libertarians who favor a traditional account of providence, it is commonly assumed that the only available option is Molinism.31 The key to the Molinist account is the doctrine of middle knowledge, according to which God knows, prior to creating, what any possible free creature would do in any set of circumstances in which it might find itself. Because the circumstances in question do not necessitate or determine what the free creature would do, what the creature freely does in those circumstances satisfies the conditions for an act’s being free in the libertarian sense. But, because God knows prior to creating what all possible free creatures would do in any set of circumstances, God can arrange for free creatures to be placed in circumstances where they will, in a libertarian sense, freely act in accord with a plan and purpose God has for each individual and for the history of the world. Middle knowledge, thus, enables God’s providence to incorporate and extend to all creaturely acts without violating libertarian freedom.32 Despite its ingenuity, there are some significant obstacles to embracing Molinism, and I suspect that even many who accept it do so only reluctantly, because they see no other way to reconcile libertarian freedom with a traditional account of providence. One problem is that the subjunctive conditionals or counterfactuals of creaturely freedom, as the objects of middle knowledge are sometimes called, are contingent propositions: in order to preserve libertarian freedom, the Molinist must hold that it is a contingent, not a necessary, matter what possible creature Jane Doe would freely do in some given possible circumstance. But we typically think that the truth-value of a contingent proposition has to be set by, or grounded in, the way things actually are. And, yet, there is nothing about the way things actually are to set the truth-value of the counterfactuals of freedom. For the Molinist denies that their truth is set by God; yet, these propositions are supposed to be true before God has created anything and so before there is anything actual other than God. Indeed, they are supposed to be true even for creatures that God never creates or true even had God decided not to create. Given that there is nothing about reality to set the truth of these contingent propositions, it has seemed to many doubtful that such propositions have a truthvalue. But, in that case, middle knowledge has no object.33 This “grounding objection” to the doctrine of middle knowledge suggests another problem, this time bearing on the requirements for libertarian freedom. As we’ve seen,

Providence, Grace, and Predestination

157

libertarians typically hold that for an act to be free it has to be ultimately up to its agent, something for which its agent bears ultimate responsibility. But an act can’t be ultimately up to an agent that doesn’t exist. And, by the same token, the truth of the counterfactual of freedom that, in certain circumstances, possible creature Jane Doe would freely agree to marry possible creature John Doe cannot be ultimately up to Jane if Jane doesn’t exist. Thus, even if the contingent propositions occupying God’s middle knowledge had truth-values, it is difficult to see why we should deem those truths ultimately up to the creaturely agents whose actions they are about. But, in that case, one wonders whether the Molinist account can preserve libertarian freedom, after all. Dual Sources reconciles libertarian freedom and traditional providence, while avoiding these problems with Molinism.34 Robust providence simply follows from the Dual Sourcer’s commitment to DUC, an implication clearly recognized by Aquinas, who explicitly bases providence on God’s universal causality: All things are subject to divine providence, not only in general, but even in their own individual selves. This is made evident thus. For since every agent acts for an end, the ordering of effects towards that end extends as far as the causality of the first agent extends. … But the causality of God … extends to all being … Hence all things that exist in whatsoever manner are necessarily directed by God towards some end. (ST 1.22.2) God is the cause not indeed only of some particular kind of being, but of the whole universal being … Wherefore, as there can be nothing which is not created by God, so there can be nothing which is not subject to His government. (ST 1.103.5) As God is the first universal cause, not of one genus only, but of all being in general, it is impossible for anything to occur outside the order of the Divine government. (ST 1.103.7) Since the very act of free will is traced to God as to a cause, it necessarily follows that everything happening from the exercise of free will must be subject to divine providence. (ST 1.22.2 ad 4)

Given DUC, free creaturely acts, like all other entities distinct from God, are in God’s hands, immediately caused by him for some purpose, whether known or unknown to us. Yet, as we saw in Chapter 4, this does not conflict with libertarian creaturely freedom. For, God’s causing our acts does not introduce any factor that determines them. Moreover, even though God has control over whether my act occurs since it cannot exist without his causing it, and since he freely causes it when he could have done otherwise, nevertheless, God’s control does not take away my own control over my act. For I perform my act voluntarily and intentionally when it was within my power voluntarily and intentionally to have done otherwise all antecedent conditions remaining the same. I even have control over God’s act of causing my act, since had I done otherwise than perform my act, God’s act, which simply consists in my act qua dependent on God, would not have occurred. We can thus accommodate the Open Theist’s desire to give us a role in shaping the future. But precisely because our free

158

Free Will and God’s Universal Causality: The Dual Sources Account

acts have dual sources—God and us—our role does not require that God step aside or entail that whatever part of the future we shape God doesn’t. The common assumption that our acts cannot be ultimately up to us if they are ultimately up to God is mistaken. Notice that Dual Sources reconciles traditional providence and libertarian freedom without any reference to middle knowledge. It should thus be attractive to theists who applaud the goals of Molinism, but worry about the grounding objection. Moreover, it arguably offers both a more secure libertarianism and a stronger account of providence than does Molinism. The libertarianism is more secure because Dual Sources avoids the worry stated above, namely, that the truths about what possible free creatures would do, on which God’s providence depends on Molinism, can’t be ultimately up to the free creatures, since the creatures don’t even exist to set these truths. The account of providence is stronger because it extends God’s control to more things, in a way that also enables Dual Sources, unlike Molinism, to embrace the traditional conception of God’s omnipotence. Consider that on a traditional conception of God’s omnipotence, God can bring about any state of affairs that is possible.35 What is more, an account of divine providence that puts in God’s hands which (if any) of two incompatible possibilities is actual will be stronger than an account on which there are some possibilities that God cannot realize. Now, on Molinism, God cannot bring about just any state of affairs that is possible and so neither is it up to God which possibilities see the light of actuality. For, on Molinism, even though it is possible that Jane turn down John’s proposal in given circumstances, God cannot create a world in which Jane does so, if the relevant counterfactual of freedom in God’s middle knowledge says that Jane would accept the proposal in those circumstances. In other words, God’s power and providence is limited on Molinism by the contents of God’s middle knowledge, over which God has no control and which restrict the range of possibilities to which God’s power extends. On Dual Sources, there is no such restriction. It is possible that Jane freely accept John’s proposal and possible that she freely turn him down, and God can bring about either one. It is, of course, not possible that God bring about Jane’s freely accepting without Jane’s accepting, when Jane could have done otherwise all antecedent conditions remaining the same. But, as we saw in Section 4.5, if Jane declines the proposal, it doesn’t follow that God could not have caused her accepting it. It only follows that God doesn’t cause her accepting it. Thus, just as Dual Sources affords a more secure libertarianism than Molinism, it also delivers a stronger conception of God’s providence and omnipotence.

8.4 Grace According to standard Christian theology, human beings after the Fall can perform good or salutary acts, at least of certain sorts, only with the help of God’s grace preceding those acts and making them possible.36 Thus stated, the doctrine of grace does not appear inimical to libertarian creaturely freedom. Granted, no human being after the Fall would enjoy the libertarian freedom to perform such a good act unless God gave the preceding grace requisite to make the act possible. But, provided God

Providence, Grace, and Predestination

159

gives that grace, and provided that the grace does not necessitate the good act but is consistent with the creature’s refraining, it appears that all the conditions for the act’s being free in the libertarian sense can be satisfied. For, in such circumstances, God’s grace would not preclude the act’s being purposeful and intentional. What is more, while God’s grace would be prior to the act, it would not be logically sufficient for the act and thus would not introduce a factor that “determines” the act in violation of libertarian freedom. Indeed, because the grace would not introduce any prior and logically sufficient condition for the act, neither would it remove the creature’s ability to do otherwise all antecedent conditions remaining the same. Nor, finally, would this grace preclude the act’s being something for which its creaturely agent is ultimately responsible, since, not necessitating the act, the grace would leave it ultimately up to the creature whether or not the act occurs.37 Although easily reconcilable with libertarian freedom, the doctrine of grace stated in the first sentence above is decidedly minimal. There are more robust accounts of grace, highly influential, on which libertarian freedom is not so easily accommodated. Consider, for example, St. Augustine, known in the western church as the “doctor of grace.” For Augustine, it is not simply that God’s grace is needed to give us the ability or even the inclination to perform good and meritorious acts. In addition, the good and meritorious acts are themselves gifts of God’s grace. Augustine takes this more robust account of grace, on which our good acts are themselves God’s gifts, from his reading of scripture. He cites certain passages as offering direct support: “We are his workmanship created in Christ Jesus in good works, which God has made ready beforehand that we may walk in them” (Eph. 2:10)38; and “For it is God who of his good pleasure works in you both the will and the performance” (Phil. 2.13)39; and “I will cause you to walk in my commandments and to keep my judgments, and to do them” (Ezek. 36:27).40 Other passages, he thinks clearly imply the account. He notes, for example, that what God commands or exhorts us to do in scripture, scripture also depicts as something God does in us. Thus, We should keep in mind that the God who tells us, “Be converted and live” (Ezek. 18:32), is He to whom we pray, “Convert us, O God” (Ps. 80:3) … We should keep in mind that He who says “Make to yourselves a new heart and a new spirit” (Ezek. 18:31) is the one who declares: “I will give you a new heart and put a new spirit within you” (Ezek. 36:26).41

Against the Pelagians, who would have us take credit for good acts that are ours and ours alone, he cites the Apostle Paul: “What hast though that thou hast not received? And if thou hast received it, why dost thou boast as if thou hadst not received it” (1 Cor. 4:7)?42 The force of Paul’s question would be severely diminished if what were received were not the good act itself, but only the ability to perform a good act, or even an antecedent resistible inclination to perform it, with the good act itself going beyond those gifts, representing value added by the creature alone. Finally, Augustine finds passages in scripture he judges difficult to reconcile without acknowledging that our good acts are God’s gifts. On the one hand, there are passages emphasizing that grace is freely given by God, not something paid as recompense for

160

Free Will and God’s Universal Causality: The Dual Sources Account

good works. Augustine, again, cites Paul: “Now to him who works, the award is not credited as a favor but as something due” (Rom. 4:4), and “If out of grace, then not in virtue of works; otherwise grace is no longer grace” (Rom. 11:6).43 On the other hand, there are passages indicating that grace is sometimes given in return for what we do. Augustine, for instance, concedes that the Pelagians are correct in reading Zechariah 1:3, “Turn to me … and I will turn to you,” as teaching that God gives grace in response to our turning to him.44 Moreover, he notes that while Paul characterizes eternal life as a grace, “The wages of sin is death, … but the grace of God is life everlasting in Christ Jesus” (Rom. 6:23; cf. Eph. 2:8–9), the Apostle also says “God will render to everyone according to his works” (Rom. 2:6), a passage in which Paul clearly has in mind the rendering of eternal life in return for good works.45 So, grace is supposed to be a free gift, a favor bestowed, rather than something due to the receiver. And, yet, grace, including the grace of eternal life, is also sometimes a recompense for good works. For Augustine, we can reconcile these seemingly contradictory claims only by recognizing that the good works for which grace and eternal life are rendered are themselves gifts of grace. Thus, Augustine maintains against the Pelagians that when God gives grace in return for our turning to him, “this turning of ours to God [is] itself also a gift.”46 And with respect to how eternal life can be at once a grace and a reward for good works, Augustine cites the passage from Ephesians quoted above: “We are his workmanship, created in Christ Jesus in good works, which God has made ready beforehand that we may walk in them” (Ephesians 2:10).47 From which Augustine concludes: If our good life is nothing more than the grace of God, then eternal life, the recompense for a good life is, without any doubt, also a grace of God; for it is freely given in recompense for that which has also been freely given. Now a good life that is so rewarded is itself simply a grace, whereas eternal life, which is given in return as a recompense, is a grace given for a grace, a kind of renumeration, as it were, in accordance with justice. Hence the truth, as it is indeed a truth, that God “will render to everyone according to his works.” (Matt. 16:27, Rom. 2:6)48

Eternal life is both a grace and a reward for good works. And it can be both because our merits are God’s gifts, so that when God awards eternal life “it is His own gifts that God is crowning.”49 There is, then, an important and influential conception of grace on which our good and meritorious acts are at once our merits and God’s gracious gifts.50 Now, one might suppose this conception works best on, or even requires, a compatibilist account of freedom. And it may be that some proponents of this conception—one could argue about Augustine—were, in fact, compatibilists. But does this conception rule out our good act’s being free in the libertarian sense, as libertarians will say they must be if we are to be genuinely responsible for them, such that they are truly meritorious? Happily, if we adopt Dual Sources, then libertarian freedom is not ruled out. But, without Dual Sources, the prospects of reconciling libertarian freedom with a robust, Augustinian (and arguably Pauline) conception of grace look much less promising.

Providence, Grace, and Predestination

161

To substantiate these points, recall from Section 7.5 that Dual Sources enables us to distinguish between divine acts that are antecedent to, concurrent or simultaneous with, and consequent to a given human action. A divine action is antecedent to a given human action if it is the bringing about of something that is temporally or explanatorily prior to that human action. A divine action is consequent to a given human action if it is the bringing about of something temporally or explanatorily subsequent to that human action. A divine action is concurrent or simultaneous with a given human action if, as understood on EM and explained in detail in Chapter 4, the divine action simply consists in the bringing about of the human action, such that both actions are logically necessary and sufficient for each other, but neither is prior to the other. As we saw in that chapter, where divine and human action is concurrent or simultaneous, the human act can be caused by God and still satisfy all the conditions for being free in the libertarian sense, including being such that its human agent could do otherwise all antecedent conditions remaining the same and such that its human agent is (together with God) ultimately responsible for the action. This concept of concurrent divine action, made possible by EM, is central to the Dual Sourcer’s reconciliation of libertarian freedom and traditional theism. But it is denied (or simply overlooked) by prevailing opinion, which holds that for God to cause a creaturely act precludes that act’s being free in the libertarian sense. The distinction between antecedent, concurrent, and consequent divine action allows Dual Sources to distinguish between antecedent, concurrent, and consequent divine grace. Each of these types of grace can refer equally either to the gift given by God or to God’s act of giving it. Antecedent grace is grace given that precedes a creaturely action, enabling or inclining the creature to perform a good or meritorious act. According to Christian theology, human beings after the Fall would not be able to perform any good acts (at least of certain sorts) without such antecedent grace. Consequent grace, by contrast, is grace given in response to a meritorious act. It is true grace—a genuine gift—because the meritorious acts, to which it is given as a response, are themselves God’s gifts. Examples of consequent grace from Augustine are God’s giving grace in response to our asking him for it and God’s rewarding eternal life for good works. In both cases, the consequent grace is a response to meritorious acts that are themselves previous gifts of grace. Finally, concurrent grace is a meritorious act given by God in an instance of concurrent divine action. The meritorious act is a grace, a gift—since it is a favor freely bestowed by God when God directly causes the act. But it is also an act that we perform, satisfying all the standard conditions for libertarian freedom.51 Now, just as prevailing opinion rejects or overlooks the possibility of concurrent divine action, so also does it reject or overlook the possibility of concurrent grace. But that makes it very difficult to see how libertarian freedom could be reconciled with the claim that our merits are through and through God’s gifts. For, clearly, we can’t explain how our merits are God’s gifts by appeal to consequent grace: consequent grace is a response to those merits. But neither does it seem possible to reconcile libertarian freedom with the claim that our merits are God’s gifts by appealing only to antecedent grace. For, either the antecedent grace is logically sufficient for the good act or it isn’t. If it is, then this grace constitutes a prior and logically sufficient condition for the

162

Free Will and God’s Universal Causality: The Dual Sources Account

act, which is to say it determines the act, contrary to libertarian freedom.52 If, on the other hand, this antecedent grace is not logically sufficient for the good act, but merely enables, or inclines us toward, the good act, then it is hard to see how the good act is fully and unequivocally something God has given. Strictly speaking, what God would have given is grace enabling the good act, or inclining us toward the good act, but not the good act itself. It would be consistent with what God gave that we not perform the good act. Whether we perform it, in the context of what God gave (and given that God does not cause the act itself), would be up to us and us alone. But this would make the good act a value we add over and above what God gives, something of credit to us that was not received from God. The problem can be illustrated by supposing two libertarian scenarios, identical in all respects including the antecedent grace given by God, but in which the creature performs the good act in one scenario, while not performing it in the other. Clearly, the creature has acted differently in these scenarios. But has God given differently? If God hasn’t, then the difference between the creature’s performing the good act and his not performing it is not made by God but by the creature alone. Hence, the creature deserves credit for something over and above what God has given; the creature has merit that is not God’s gift. Against the foregoing argument that we can’t explain by appeal to antecedent grace alone how a meritorious act is both free in the libertarian sense and wholly God’s gift, two rebuttals might be attempted. First, it might be urged that even if the antecedent grace is not logically sufficient for the good act, and even if the good act is not caused by God, we could still say that the good act is fully God’s gift. For it might be argued that even while not by itself logically sufficient for the good act, the antecedent grace once placed in the soul by God is logically sufficient on the condition that the creature not resist the grace. Since the antecedent grace given by God will necessarily result in the good act provided the creature simply not resist, it might be maintained that, in a case where the creature doesn’t resist, the good act has been fully given by God in virtue of the antecedent grace. Yet, it is doubtful that this first rebuttal succeeds. For starters, what does the creature’s not resisting the antecedent grace consist in? If it consists simply in his performing the good act, then, of course, the antecedent grace coupled with this nonresistance is logically sufficient for the good act. But, at least for those libertarians who think that the good act is caused by and up to the creature alone, the same problem arises as before: the good act will be something added by the creature over and above what God has given. We need not consider in detail other possibilities for what the creature’s nonresistance might consist in. For, whatever it consists in, if the nonresistance is up to the creature alone, then the same problem arises. The antecedent grace given by God will not by itself be sufficient for the good act. On the contrary, it will be up to the creature alone whether or not the antecedent grace God has given results in the creature’s performing the good act. But, then, the creature’s performance of the good act is something over and above what God has given, and the creature deserves credit for something beyond what she has received. A second rebuttal, in contrast to the first, allows that the antecedent grace is all by itself logically sufficient for the good act. It, thus, would enable us to say that the good act simply follows the grace once it is given. Against the worry that logically

Providence, Grace, and Predestination

163

sufficient antecedent grace would rule out libertarian freedom, this second rebuttal maintains that God can’t give the antecedent grace in the first place (i.e., can’t infuse it into the creature’s soul) unless the creature refrains from resisting or setting up obstacles to God’s giving that grace, an omission of resistance that is up to the creature alone. Because the good act that follows necessarily from the antecedent grace will not happen without the prior nonresistance of the creature that is up to the creature alone, the rebuttal maintains that the good act that follows is sufficiently under the control and responsibility of the creature to count as free, at least, in the broad libertarian sense. On this proposal, the creature is derivatively responsible for the good act that follows necessarily from the antecedent grace in virtue of being responsible for the omission of resistance, which is up to the creature alone and which enables God to give the antecedent grace. Like the first rebuttal, it is doubtful that this second rebuttal succeeds. There are three distinguishable problems. First, on this proposal, the creature could be morally responsible for the good act that follows from the antecedent grace only if she is morally responsible for the prior nonresistance that enables God to give the antecedent grace, that is, only if the prior nonresistance is an exercise of free will of the sort that constitutes an exercise of responsibility and control. But if the prior nonresistance is an exercise of free will of this sort, then, given that the nonresistance is certainly a good exercise of will, it looks as if we should recognize the nonresistance as something meritorious. But the nonresistance is not God’s gift, since, on this second rebuttal, it is up to the creature alone. Thus, the rebuttal still leaves us with a creaturely merit that is not God’s gift, though this time the non-gifted merit comes before God gives the antecedent grace, not after, as on the first rebuttal. Second, since the creature’s nonresistance on this second rebuttal is the decisive first step that decides whether God can even give the grace that results in the good act, the second rebuttal would appear to deny God’s freedom to give good acts to whomever he wants, in conflict with what is normally maintained by those who hold that our merits are God’s gifts.53 Finally, we need to ask, “Is the very first antecedent grace that God gives a particular creature one that God can give only with the creature’s prior nonresistance, up to the creature alone?” If the answer is “yes,” then it would appear that the creature has the initiative with respect to its own life of grace, in a way that smacks of Semi-Pelagianism. If the answer is “no,” then the good act for which this first antecedent grace is logically sufficient will not be up to the creature, since the second rebuttal takes the creature’s responsibility for the good act that follows from an antecedent grace to be derivative on the creature’s prior nonresistance.54 Dual Sources eliminates these problems that emerge from trying to reconcile libertarian freedom with the claim that our merits are God’s gifts by appeal to antecedent grace alone. On Dual Sources, in addition to the antecedent grace that enables or inclines the creature toward the good act, there is also concurrent grace: the giving of the good act itself, caused by God in an act that is simultaneous with the good creaturely act and that is therefore consistent with that act’s being free in the libertarian sense for reasons already explained. Though free in the libertarian sense, the good act performed by the creature is, in its entirety, given by God. There is no good introduced by, and creditable to, the creature alone. If God gives the exact same antecedent grace in two identical scenarios, and if the creature performs the good act in one scenario but

164

Free Will and God’s Universal Causality: The Dual Sources Account

not in the other, God will have given something in the one that he did not give in the other—the good act itself. We can thus boast of no good that has not been given from on high. Any merit we enjoy is, without remainder, God’s gift.55 As far as I can tell, only Dual Sources can succeed—or, at least, it is the view that most clearly succeeds—in reconciling libertarian freedom with a robust, Augustinian conception of grace. To hold that antecedent grace is not logically sufficient for a good creaturely act, while holding that the performance of the good act in light of that grace is up to the creature alone, is to forfeit the claim that the good act is God’s gift. To hold that the antecedent grace is logically sufficient for the good act would explain how the act is God’s gift, but it would do so in violation of libertarian freedom, unless it is up to the creature alone whether God can give the antecedent grace in the first place. But to hold that it is up to the creature alone whether God can give the antecedent grace in the first place is to introduce a non-gifted creaturely merit prior to God’s giving of antecedent grace, to confer on the creature the initiative in its own life of grace, and to limit God’s freedom with respect to whom he can give good acts. What’s needed is an account on which antecedent grace does not determine or necessitate the good act, yet on which (in what I have called concurrent grace) the good act is caused by God according to a model of divine agency on which its being caused by God is consistent with its being free in the libertarian sense. I know of no account other than Dual Sources that provides what’s needed.56 Before closing this section, it will be useful to comment a bit further on Dual Sources’ understanding of grace and then to consider an objection. Augustine sometimes speaks of the will as being “unwaveringly and invincibly influenced by divine grace,”57 language that may suggest that God’s grace is irresistible. Leaving the proper interpretation of Augustine to one side, and assuming the failure of the second rebuttal discussed above, if we take “irresistible grace” to be grace that is logically sufficient for a good act, then we have already seen that irresistible antecedent grace is not consistent with the good act’s being free in the libertarian sense. For, antecedent grace is prior to the good act, so that if it is also logically sufficient for the act, then it determines the act contrary to libertarian freedom and removing the creature’s ability to do otherwise all prior conditions remaining the same. Nevertheless, the Dual Sourcer can happily admit that concurrent grace is irresistible. For, concurrent grace refers either to the good act itself that God causes by a concurrent divine action or to God’s concurrent act of causing the good act. And both of these are logically sufficient for the good act. It is not possible that the good act, or God’s act of causing the good act, obtain without the good act’s obtaining. The irresistibility of concurrent grace is consistent with libertarian freedom because, as explained in detail in Chapter 4, given EM, God’s act of causing the good act simply consists in the good act qua dependent on God, and thus God’s act is not prior to the good act. Accordingly, God’s act neither determines the good act nor takes away the creature’s ability to do otherwise all prior conditions remaining the same. To be sure, it is not possible that God cause the good act and the creature not perform that act. But whether or not God causes the good act—whether he gives concurrent grace—is not entirely outside the creature’s control. For, the creature could have done otherwise in those very same antecedent conditions, and had the creature done

Providence, Grace, and Predestination

165

otherwise, God’s act of concurrent grace would not have occurred. By the same token, as we saw in Section 4.5, since the creature’s act is not prior to God’s act, in a case where the creature does not perform the good act, it does not follow that God could not have caused the creature’s performing it, all prior conditions remaining the same. Thus, while the irresistibility of concurrent grace does not entail that the creature could not have refrained from the good act all prior conditions remaining the same, neither would the creature’s not performing the good act entail that God wasn’t able freely to bestow the concurrent grace. To sum up, given Dual Sources, antecedent grace, which enables or inclines the creature to the good act, is prior to that act without being logically sufficient; and concurrent grace, in which God causes the good act, is logically sufficient to the good act (and, so, irresistible) without being prior. Thus, neither type of grace determines the creaturely act or removes the creature’s ability to do otherwise. And, yet, nothing the creature does (or doesn’t do) limits God’s ability freely to bestow these graces.58 Now, the objection. Granting that Dual Sources shows how our good acts can be God’s gifts and yet be free in the libertarian sense, the accomplishment might be thought considerably diminished once it is noticed that, on Dual Sources with its commitment to DUC, our bad acts are likewise caused by God in instances of concurrent divine action. Must we say, then, that our bad acts are also God’s gifts? And, if so, does that admission nullify the advantage that Dual Sources can claim in being able to reconcile libertarian freedom with a robust, Augustinian conception of grace? An initial response might note that the relevant sense of “gift” connotes something that is good for the recipient. Now, a Dual Sourcer could say that for any given sin, falling into that sin either is ultimately good for the sinner and permitted by God as such or is not ultimately good for the sinner. If it were ultimately good for the sinner and permitted by God as such, then God’s causing the act of sin would be, in a way, a kind of gift or grace, and it would be no disadvantage to admit as such. On the other hand, if the sin is not ultimately good for the sinner, then it does not count as a gift, and so we needn’t, as in the case of our good acts, call these bad acts “gifts,” even though both types of acts are caused by God. The foregoing response turns on the meaning of the word “gift,” but it might be argued that the real problem remains whether or not that particular word applies. The real problem, one might argue, is just that, on Dual Sources, our acts of sin are caused by God every bit as much as our good acts. Now, presumably, the ideal would be an asymmetrical account on which our good acts are caused by God while our bad acts by us alone. But, if the choice is between an account on which we can say God causes our good acts but at the cost of having to say he causes our bad, and an account on which we can deny that God causes our bad acts but at the cost of having to deny that he causes our good, why think the former option is to be preferred? Well, one might think it is to be preferred because scripture seems to prefer it. As we’ve seen at various points in this book (e.g., Sections 2.1 and 7.6), scripture presents God as causally behind not only our good actions but (in multiple passages) our bad as well. That said, it would be a mistake to think that, on Dual Sources, God causes sins just as he causes our merits. On the contrary, at least if the argument of Chapter 6 succeeds, Dual Sources makes possible an attractive asymmetry, whereby we can

166

Free Will and God’s Universal Causality: The Dual Sources Account

affirm that God causes our merits while also denying that he causes our sins, and this despite the fact that, given DUC, we must acknowledge that God causes the act of sin. Given the argument of Chapter 6, a sin of action consists of two elements: the act and the privation of conformity to the moral standard in virtue of which the act is sinful and in which the act’s sinfulness consists. In order to cause the sin, one must cause both the act and the privation. But while God causes the act, only the sinner causes the privation as well as the act, and thus only the sinner causes the sin. Indeed, as we saw in Chapter 6, the act God causes is an exercise of active power and all exercises of active power are, as such, good. So, what God causes, even in an act of sin, is something good as such, even if we can also denominate it bad in virtue of its privation, which God does not cause. It is true, then, that, given DUC, God causes our acts of sin just as he causes our good acts. But to cause an act of sin is not to cause sin itself. And, so, on Dual Sources, the ability to affirm that God causes our good and meritorious acts does not come at the price of having to affirm that God causes sin.

8.5 Divine–Human Dialogue Critics of theological determinism sometimes charge that the view undermines the possibility of genuine divine–human dialogue. William Alston, for example, maintains that if theological determinism is true, then ostensible dialogue between God and human beings is really just a very complicated monologue, a conversation God has with himself, like the conversation between a ventriloquist and his dummy or between a hypnotist and a patient whose every word is controlled by the hypnotist’s suggestions.59 Now, clearly, the ventriloquist and hypnotist analogies fail to capture divine–human dialogue on Dual Sources. It is true on Dual Sources that God causes all the acts that make up the human being’s side of the conversation. But, unlike the ventriloquist’s dummy, the human being on Dual Sources is very much alive and capable of literal speech. And, whereas the pretend words of the dummy and the real words of the hypnotist’s patient are (presumably) determined by antecedent and logically sufficient conditions introduced by the ventriloquist and the hypnotist, God’s bringing about our side of the conversation on Dual Sources leaves our contribution ultimately up to us and something of which we could have done otherwise all antecedent conditions remaining the same. Still, even if, unlike theological determinism, Dual Sources is consistent with our making a contribution to the conversation that is free in the libertarian sense, one might think that the control God has over our side of the conversation is incompatible with genuine dialogue. According to Alston, “dialogue requires two independent participants, neither of which wholly controls the responses of the other.”60 But, on Dual Sources, God has control over my responses, since (a) he causes them, (b) his causing them is both logically necessary and logically sufficient for my so responding, and (c) he could have done otherwise than cause them. Of course, on Dual Sources, it is not possible that God cause my responses unless I respond when I could have done otherwise all antecedent conditions remaining the same. I, thus, have control

Providence, Grace, and Predestination

167

over God’s acts of causing my responses. But God has control over my responses, all the same, and that control appears to be incompatible with what Alston thinks genuine dialogue requires. “Divine–human dialogue” constitutes a particular type of the more general category of “divine–human interaction.” Broadly speaking, “divine–human interaction” refers to God and human beings acting in light of, or in response to, actions directed by each to the other. “Divine–human dialogue” is simply divine–human interaction that involves God and human beings in conversation. In the remainder of this section, I will show, first, how Dual Sources can accommodate divine–human interaction in general and then return to the question of whether Dual Sources undermines particular requirements for divine–human dialogue. What do we need to make possible divine–human interaction? Simply that God and human beings can do things in light of, or in response to, what the other has done. Now, such interaction is certainly possible given Dual Sources. Recall the distinction from Section 7.5 and the previous section between divine actions that are antecedent to, concurrent or simultaneous with, and consequent to a given human action. Just as we can speak of antecedent, consequent, and concurrent divine actions, relative to a given human action, so we can speak of antecedent, consequent, and concurrent human actions relative to a given divine action. A human action is antecedent to a given divine action if it is prior in the order of explanation to that action or if it is temporally prior to what is brought about in the divine action to which it is antecedent. A human action is consequent to a given divine action if it is subsequent in the order of explanation to that action or if it temporally follows what is brought about in the divine action to which it is subsequent. A human action is concurrent with a given divine action if it is brought about in that divine action, such that the divine and human actions are logically necessary and sufficient for each other, but neither is prior to the other. With the foregoing in place, we can see that Dual Sources easily accommodates divine–human interaction. It allows for God and human beings to act in light of, or in response to, what the other does. For example, God can issue a command, to which I can respond, and then God can respond to my response. Or I can pray for God’s help and God can answer, either by granting my request or by declining for his good purposes. Within the “flow” of such interaction there will be, relative to one of my actions, divine actions that are antecedent, consequent, and concurrent, just as my various actions will be antecedent, consequent, or concurrent relative to one of God’s actions. For example, God’s command is antecedent to my acting in response to that command. God’s response to my response is consequent to my response. God’s causing my act in response to his command is concurrent with my act, as my act is concurrent with God’s causing it. And so on. For one of God’s actions to constitute a response to one of my actions is simply for God to perform that action because of, or in light of, my action. In such a case, my action enters into the reason for God’s performing the action that constitutes his response. Notice that distinguishing between antecedent, consequent, and concurrent divine acts is perfectly consistent with divine simplicity, immutability, and eternity. Given EM, all of God’s acts of bringing things about and so also all of God’s contingent

168

Free Will and God’s Universal Causality: The Dual Sources Account

willings are extrinsic to him, and therefore a diversity of such acts does not violate divine simplicity by implying a diversity of intrinsic items in God. Neither does God’s performing a diversity of such acts, some of which are antecedent relative to a given human act, some of which are consequent, require that God undergo intrinsic change. For, since these divine acts are extrinsic to God, the fact that one act is before the other—whether explanatorily or temporally—does not imply any real change in God. We can even date God’s actions, if we want, without compromising God’s eternity. Proponents of God’s eternity often maintain that there is no before or after in God’s action; rather, they say that God’s action is eternal and that what is before and after are only the effects of that action.61 On this approach, we cannot say that God parted the Red Sea before he gave the Ten Commandments. We can only say that God brought it about that the Red Sea opened up before Moses received the Commandments. The assumption that if God is eternal, then his action must be eternal (and so not dateable) follows from assuming that God’s action is intrinsic to God and so must have the same eternal mode of existence that God has. But, given EM, God’s causal acts are not intrinsic to God. Rather, they consist in God’s effects together with the causal-dependence relations those effects have to God. Since these effects, which at least partially constitute God’s acts, have a temporal location, we can also temporally locate God’s acts of bringing them about. We can say very naturally that God parted the Red Sea before he gave the Ten Commandments. Yet, since the subject of these acts exists eternally, we can also say that, from all eternity, God parted the Red Sea before he gave the Ten Commandments. There is no contradiction in speaking this way since the locution “from all eternity” notes the vantage or “place” from which God acts, even while the actions themselves share the temporal location of the effects that partially constitute them. Dating God’s causal actions does not violate the essential claims of eternity, namely, that God has no beginning or end and that there is no succession in God. Note further that, as discussed in Section 8.2, all of God’s causal acts, whatever their temporal location, are present to God’s eternity in virtue of the fact that all of time is present to God’s eternity. And, since God knows his creatures in his acts of bringing them about, God knows from all eternity all that exists at any time, even though he knows these things in virtue of causal acts that can be assigned temporal locations. Having seen that Dual Sources enables us to accommodate divine–human interaction in general, what about the Alston-inspired worry that Dual Sources might preclude divine–human dialogue? As we’ve seen, Alston contends that “dialogue requires two independent participants, neither of which wholly controls the responses of the other.”62 Why? Because he thinks each party to a genuine dialogue must satisfy two conditions63: 1. Each party must be ultimately responsible for his or her side of the dialogue, in the sense of initiating and authoring that side; and 2. Each party must not be responsible for the other party’s side of the dialogue, but must rather confront and respond to the other’s contribution as something independent of his or her will and outside his or her control.

Providence, Grace, and Predestination

169

Dual Sources has no trouble accommodating condition (1) when applied to divine– human dialogue. God’s side of the conversation is ultimately up to God. And, as we have noted, God’s causing our side of the conversation does not preclude our side’s being ultimately up to us, something thoroughly within our control. Yet, it seems doubtful that Dual Sources can accommodate condition (2). For while the human party to a divine–human dialogue is certainly not responsible for God’s side of the conversation, God, since he causes our actions, clearly is responsible for the human side. Now, Alston concedes a sense in which the Dual Sourcer could say that God “replies” to the human side of the conversation. For, he admits even for the theological determinist that God could say certain things because of, and therefore in reply to, what the human party to the conversation said—including things that God would not say had the human party spoken differently.64 Nevertheless, says Alston: We still have to say that this fails to be a reply in the full-blooded sense that is required for “genuine dialogue”. What, then, is missing? Just this. [That which is] putatively responded to in no way “stands over against” God as something independent of His will, something introduced into the situation by the initiative of another, something to which He has to adjust His conduct, something that requires a special ad hoc “response” on His part. In “replying” to Moses’ question God is merely adjusting one decision of His to another. The dynamics of the affair are wholly internal to the divine conation. He is not confronted with something to which He has to fashion a response. … Without an other that is sufficiently outside my control to make its own independent contribution to what is going on, there is nothing for me to reply to.65

What should the Dual Sourcer make of this challenge? First, in accord with points already made, some of what Alston says (which is, after all, directed against theological determinism) does not hold for Dual Sources. On Dual Sources, for example, what God responds to is “something introduced into the situation by the initiative of ” the human party; the human side of the dialogue is ultimately up to the human being as much as it is to God. For the same reason, it is not true that, on Dual Sources, “the dynamics of the affair are wholly internal to the divine conation.” For God’s willing or purposefully bringing about the human side of the conversation would not even have occurred without the human party’s purposefully making her contribution, when she could have done otherwise all antecedent conditions remaining the same. Far from the dynamics of the affair being wholly internal to the divine conation, that part of the divine conation that consists in God’s concurrent causal-willing of the human side of the dialogue is not outside the control of the human conation. Still, it is true, on Dual Sources, that God does not confront the human side of the conversation as something independent of God’s will and outside his control. But is this a liability of Dual Sources? I would suggest the contrary. Another name for the human side of the divine–human dialogue is “prayer.” And, at least for a main stream of the Christian tradition, prayer, while being something we do, is also, and importantly, something God actively does in and for us. Herbert McCabe speaks for this tradition when he remarks:

170

Free Will and God’s Universal Causality: The Dual Sources Account

For traditional theology, prayer is not our attempt to gain the Father’s attention, prayer is not in fact primarily a human activity, it is something we do in virtue of being divine, it is, to use the traditional language, the work of grace in us … It is God who prays. Not just God who answers prayer but God who prays in us in the first place. In prayer we become the locus of the divine dialogue between the Father and the Son … My prayer is the action of God’s grace in me. My prayer is not me putting pressure on God, doing something to God, it is God doing something for me, raising me into the divine life or intensifying the divine life in me.66

Thus, the Second Council of Orange (529), in response to the Semi-Pelagians, rejects holding that “the grace of God can be conferred because of human prayer, and not … that it is grace itself that prompts us to pray.”67 Augustine insists that “this is also of the divine gift, that we pray; that is, that we ask, seek, and knock.”68 Aquinas agrees that “the very act of praying is a gift of God.”69 So, too, does Karl Barth: “God … causes our prayer to proceed from his grace.”70 St. Paul writes that “when we cry, ‘Abba Father!’ it is that very Spirit bearing witness with our spirit that we are children of God” (Rom. 8:15–16). The suggestion seems to be that the Holy Spirit is at work when we call out to God. Consider, also, the Apostle’s remarks later in the same chapter of Romans: Likewise the Spirit helps us in our weakness; for we do not know how to pray as we ought, but that very Spirit intercedes with sighs too deep for words. And God, who searches the heart, knows what is the mind of the Spirit, because the Spirit intercedes for the saints according to the will of God. (Rom. 8:26–27)

Commenting on this admittedly hard-to-interpret passage, Augustine says that “the Holy Spirit, who intercedes with God on behalf of the saints, … moves us to pray when we groan, and thus he is said to do what we do when he moves us.”71 Christian philosopher Paul K. Moser, with this passage in mind, characterizes prayer as “the gift of God’s Spirit communing with God through us.”72 New Testament scholar Joseph A. Fitzmeyer concludes from this passage that “it was part of [God’s] plan of salvation that the Spirit should play … a dynamic role in the aspirations and prayers of Christians.”73 St. Teresa of Avila gives poetic expression to the conviction that prayer is God’s gift, as she addresses her Lord: Give me, if You will, prayer; Or let me know dryness, An abundance of devotion, Or if not, then barrenness.74

Similarly, Julian of Norwich writes of petitionary prayer that God “moves us to pray for what it pleases him to do, and for this prayer and good desire that come to us by his gift he will repay us.”75 And of contemplative prayer, she writes, “for the whole reason why we pray is to be united into the vision and contemplation of him to whom we pray,

Providence, Grace, and Predestination

171

wonderfully rejoicing with reverent fear, and with so much sweetness and delight in him that we cannot pray at all except as he moves us at the time.”76 Finally, this understanding of God’s agency within our prayer finds reflection in the Liturgy of the Hours, whose invitatory, commencing the first hour each day, highlights God’s active role: “Lord open my lips … and my mouth will proclaim your praise.”77 It likewise finds reflection in the Catechism of the Catholic Church, which describes prayer as “the action of God and of man, springing forth from both the Holy Spirit and ourselves.”78 Alston takes as his model of “genuine dialogue” the ordinary relationship between two human beings in conversation with one another, a relationship where each party confronts the contribution of the other as something offered independently of his will and agency. But if the foregoing understanding of prayer is correct, then divine– human dialogue cannot be like that. If prayer is a gift from God, then it is something God brings about in us. And if it is something God brings about in us, then it is not apart from God’s will and control. While Dual Sources cannot accommodate divine–human dialogue of the sort that Alston envisions, it can accommodate very well the tradition that understands divine–human dialogue, or prayer, to be God’s gracious gift. What is more, it can accommodate this tradition alongside the affirmation that our prayers are, at the same time, ultimately up to us and fully free in the libertarian sense. On Dual Sources, we can affirm without contradiction both that we pray to God in acts that satisfy all the conditions for libertarian freedom and that we should thank God for giving us these very acts by which we are drawn into God’s life and friendship. I know of no account other than Dual Sources that better supports both these affirmations.79

8.6 Predestination The Westminster Dictionary of Christian Theology defines predestination as “the doctrine that God foreknows, and ordains, from all eternity, who will be saved.”80 The Catholic Encyclopedia states that “the notion of predestination comprises two essential elements: God’s infallible foreknowledge (praescientia), and His immutable decree (decretum) of eternal happiness,”81 which latter refers to God’s choice or election of those persons who reach eternal happiness. That, on Dual Sources, God knows who will attain salvation follows from the discussion in Section 8.2. In this section, I will use “predestination” to refer exclusively to the second element of the foregoing definitions. I will be interested in a strong conception of predestination understood as the following conjunctive claim: Predestination: Necessarily, a created person attains salvation if and only if God chooses that that person attain it, and, for any possible created person, it is within God’s power (whether or not) to choose that that person attain salvation.

Certainly, many giants of the Christian tradition have affirmed the substance of the foregoing claim, and it should not surprise us that they have done so.82 For, although

172

Free Will and God’s Universal Causality: The Dual Sources Account

the proper interpretation of Christian scripture on the issue is a matter of controversy, it is, at the very least, not implausible to read scripture as affirming or presupposing Predestination.83 Notice that, given Predestination, the salvation of any created person is something that it is within God’s power to deliver. Accordingly, it makes sense to beseech God that we attain salvation, a prayer that is common to much Christian practice and liturgy. Consider, for example, this prayer offered by the priest in the Catholic Mass: Therefore, Lord, we pray: graciously accept this oblation of our service, that of your whole family; order our days in your peace, and command that we be delivered from eternal damnation and counted among the flock of those you have chosen.84

Or consider a prayer such as the following from the Liturgy of the Hours: Lord God, … keep us faithful to your law in thought, word and deed. Be our helper now and always, Free us from sin, and bring us to salvation in that kingdom where you live and reign with the Father and the Holy Spirit, one God, for ever and ever.85

Such prayers for the attainment of salvation would make little sense were the attainment outside God’s hands—for example, if God were only able to extend us the offer of salvation and some help in pursuing it, but did not also have control over all that must happen in order for our salvation to be realized. Given Predestination, the attainment of salvation—and not just the offer or possibility of it—is God’s gracious gift, something for which we should pray and something for which, if granted, we should be grateful.86 As with the conceptions of providence, grace, and prayer discussed in previous sections, however, my aim is not to establish the truth of Predestination. Rather, my aim is to show how Dual Sources can accommodate it, alongside a libertarian understanding of a second claim: Responsibility: Whether a person attains salvation depends on what that person does.

Like Predestination, one may reasonably judge Responsibility to be affirmed in scripture,87 and, thus, Responsibility also has strong support within the Christian tradition.88 But, at least on the surface, it may appear difficult to reconcile the two claims, where what the person does as necessary to attain salvation is understood to be free in the libertarian sense. Surely, Oliver Crisp represents a widely held view, when he suggests that “libertarianism is not consistent with the notion that God ordains and brings about (i.e. ensures) the salvation of anyone.”89

Providence, Grace, and Predestination

173

In thinking about this issue, it will be helpful to have some way of referring to or characterizing the creaturely act (or acts) on which a person’s attaining salvation depends. To accommodate a range of views on the matter, I will simply refer to this act (or these acts) as the creature’s “accepting the offer of salvation,” where “accepting the offer” serves as a kind of placeholder for whatever it is that the creature must do in order to attain salvation, whether that be believing, performing good works, repenting, some combination of these, or something else. Why do Predestination and a libertarian reading of Responsibility appear in conflict? Well, given Predestination, God’s choice that a person attain salvation is clearly logically sufficient for that person’s attaining: The person attains salvation if and only if God chooses that that person attain it. But, given Responsibility, a person attains salvation only if that person accepts the offer of salvation. It follows that, given both Predestination and Responsibility, God’s choice that a person attain salvation is logically sufficient for the person’s accepting the offer: It is not possible that God make the choice and the person not accept. And, yet, given Predestination and Responsibility, one might think that God’s choice that a person attain salvation is also causally or explanatorily prior to the person’s accepting. For, otherwise, it might seem to be outside God’s power whether a person performs the act(s) that Responsibility holds necessary for that person to attain salvation. And, of course, were it is outside God’s power whether a person performs the act(s) needed, then, contrary to Predestination, God would not have the power, for any possible created person, to choose that that person attain salvation such that God’s choice is logically necessary and sufficient for the person’s attaining it. It may seem, then, that God’s choice that a person attain salvation is both prior to and logically sufficient for the person’s accepting the offer of salvation. But, in that case, God’s choice will constitute a factor that “determines” the creature’s accepting in the sense of “determines” that rules out libertarian freedom. Moreover, if God’s choice is both prior to and logically sufficient for the person’s accepting, the person could not have done otherwise than accept, all antecedent conditions remaining the same: There would be no possible world in which God’s prior choice that the person attain salvation obtains and yet the person does not accept the offer. Finally, assuming that if God’s choice is explanatorily prior, then God’s choice is outside the person’s control; it would seem that the person’s accepting the offer was not ultimately up to that person, since the accepting results necessarily from a factor (God’s choice) over which the person has no say. Confronted with this difficulty, a Molinist might try to reconcile Predestination with a libertarian reading of Responsibility using the doctrine of middle knowledge. On the Molinist proposal, God knows prior to creating whether a particular person would accept the offer of salvation in any possible set of circumstances in which the person might find herself. In light of this knowledge, God can choose that certain people attain salvation by placing them in circumstances in which he knows they will accept the offer. If we assume that there are circumstances in which any possible created person would accept the offer, then, in keeping with Predestination, it would be within God’s power, for any possible created person, whether to choose that that person attain salvation; and the choice would be logically sufficient for the attaining since it would be made in

174

Free Will and God’s Universal Causality: The Dual Sources Account

light of certain knowledge that the person would accept the offer in the circumstances. But, claims the Molinist, the proposal can also accommodate a libertarian reading of Responsibility. For the antecedent conditions would leave it possible for the person not to accept the offer of salvation in those very same circumstances in which, by his middle knowledge, God knows that the person would in fact accept. The Molinist proposal understands a person’s accepting the offer of salvation to be, in a way, prior to God’s choice that the person attain salvation. For, on this proposal, God makes his choice regarding a person’s attaining salvation in light of his logically prior middle knowledge regarding whether the person would accept the offer in various circumstances. Thus, Molinists have sometimes said that God’s predestining is conditioned by a prevision of merits, a prior knowledge of whether the creature would accept the offer of salvation.90 Here, let me simply make three points regarding the Molinist proposal for reconciling Predestination with a libertarian reading of Responsibility. First, it depends on the viability of Molinism, more generally. As we saw in Section 8.3, the grounding objection and the worry that Molinism doesn’t really accommodate our acts being ultimately up to us will be obstacles for some would-be Molinists. Second, the Molinist proposal would not appear consistent with the robust conception of grace discussed in Section 8.4. For, a person’s accepting the offer of salvation is, presumably, a good and meritorious act. But, on Molinism, whether the creature would accept the offer of salvation in a given set of antecedent conditions (including antecedent grace) is up to the creature alone. And, so, contrary to the robust conception of grace, the creature will perform a good act that is not without remainder God’s gift, deserving credit for something beyond what has been received from God. Third, even granting the viability of Molinisim, whether the Molinist proposal succeeds in reconciling Predestination with a libertarian reading of Responsibility depends on the content of God’s middle knowledge. One might suppose that for any possible created person, there are at least some possible circumstances in which that person would accept the offer of salvation. But, if there are possible created persons who, God knows by middle knowledge, would not accept the offer of salvation in any possible circumstances, then, given Responsibility, God could not choose that such persons attain salvation in a way that ensures that they will attain it. And that result conflicts with Predestination. Thus, if God’s middle knowledge includes knowledge of possible persons who would, under no possible circumstances, accept the offer of salvation, then the Molinist proposal won’t enable a reconciliation of Predestination and a libertarian reading of Responsibility, after all, though it might enable a reconciliation with a different (likely weaker) conception of predestination.91 Unlike Molinism, Dual Sources denies any sense in which the creature’s accepting the offer of salvation is prior to God’s choosing that the creature attain it. Yet, Dual Sources also denies that God’s choosing is prior to the creature’s accepting. For this reason, contrary to initial appearances, God’s choosing that a person attain salvation does not constitute a prior and logically sufficient condition that “determines” the creature’s accepting, in conflict with libertarian freedom. What exactly is God’s choosing that a person attain salvation, on Dual Sources? To prepare for an answer, let us stipulate that a person’s accepting the offer of salvation

Providence, Grace, and Predestination

175

presupposes an antecedent divine act (or acts) of offering the person salvation and giving graces that enable or incline the person to accept the offer, but which graces are consistent with the person’s not accepting the offer all antecedent conditions remaining the same. Let us stipulate, also, that God makes this offer of salvation and gives these graces to all created persons. Given Responsibility, a person must accept the offer in order to attain salvation. And, let us stipulate, finally, that—perhaps because of God’s goodness or of some essential connection between accepting and attaining—it is not possible for a person (at least finally) to accept the offer of salvation and not attain it. On Dual Sources, then, God’s choice that a person attain salvation just is his bringing about the person’s accepting the offer in order that the person attain it. Crucially, on this view, God’s act of bringing about the person’s accepting the offer, and thus God’s choice that the person attain salvation, is not antecedent to the person’s accepting. Rather, God’s act and choice is concurrent with the person’s accepting. It is an instance of concurrent divine action, a type of divine action that, as we have seen, allows the person’s act of accepting to satisfy all the standard conditions for libertarian freedom: lack of determination, the ability to have done otherwise all antecedent conditions remaining the same, and the accepting’s being ultimately up to the created person. Once the creature has accepted the offer of salvation, Dual Sources also allows for a consequent divine act of God’s rewarding salvation to those who have accepted it: “Come, you that are blessed by my Father, inherit the kingdom prepared for you from the foundation of the world” (Matt. 25:34).92 As we saw above, it might be thought that God’s choice that a person attain salvation has to be prior to the person’s accepting the offer, if it is really going to be within God’s power, for any created person, whether to choose that that person attain salvation. But, on Dual Sources, God’s choice that a person attain salvation just is God’s bringing about the person’s accepting the offer in order that the person attain it. And, as we saw in Sections 4.4–4.5, God’s act of bringing about a creaturely act A can be concurrent or simultaneous with A, and yet it still be that God has it within God’s power to bring about A or not, all antecedent conditions remaining the same. On Dual Sources, then, the divine power implied in Predestination holds consistent with denying that God’s choice that a person attain salvation is prior to the person’s accepting the offer. Dual Sources secures a libertarian reading of Responsibility because it accommodates the claim that a person’s attaining salvation depends on what that person does, and does freely in the libertarian sense, having the ability to do otherwise all prior conditions remaining the same. Because a person attains salvation if and only if she accepts the offer, and because it is ultimately up to the person whether she accepts once the offer is made and the antecedent graces given, it is ultimately up to the person whether she attains salvation. Yet, Dual Sources at the same time accommodates Predestination, with its implication that it is ultimately up to God whether the person attains salvation. Dual Sources, thus, enables us to say that whether a created person attains salvation is ultimately up to both God and that person. But doesn’t the very word “predestination” suggest that, contrary to the foregoing proposal, God’s choice that a person attain salvation is an antecedent decree, a divine act that is explanatorily prior to the person’s doing what is required to attain salvation? To think so is not implausible, but neither is it necessary. Just as one may plausibly hold

176

Free Will and God’s Universal Causality: The Dual Sources Account

that the essential feature of a commitment to divine foreknowledge is simply that God knows future events, and not that God or God’s knowledge is prior to the future events God knows, so I maintain that the essential feature of predestination is that a person’s attaining salvation is God’s choice and within God’s power, and not that God’s choice is prior to the person’s doing what is needed to attain salvation. If one thought that the priority of God’s choice to the creature’s accepting were essential, then one would need to modify Predestination to include that priority. Suffice to say that such a modified understanding would be incompatible with a libertarian reading of Responsibility. Note, also, that Predestination left unmodified, and as consistent with Dual Sources and a libertarian reading of Responsibility, is sufficient to make intelligible the claim that the attainment of salvation is a gift from God—that we should pray for the attainment of salvation for ourselves and others, and that we should thank God if the attainment is granted. For even though Predestination and Dual Sources allow for a libertarian reading of Responsibility, they also make it clear that a person attains salvation if and only if God chooses that the person attain it, and that it is within God’s power whether or not so to choose. The foregoing shows how Dual Sources enables us to reconcile Predestination with a libertarian reading of Responsibility. But is Dual Sources committed to Predestination? If so, is that a good thing? It seems to me that Dual Sources likely is committed to Predestination. First, Dual Sources embraces DUC, and it follows from DUC not only that God brings about a creature’s accepting the offer of salvation but that God brings about whatever exists when creatures occupy whatever state they occupy, whether that state be (or be in) Heaven, Hell, Purgatory, the state of the wayfarer, or any other state one proposes. Thus, in virtue of DUC, Dual Sources appears committed to the first conjunct of Predestination, which holds that “necessarily, a created person attains salvation if and only if God chooses that that person attain it.” Of course, Predestination holds more than that God brings about existence in Heaven, for those who exist there. It also holds that “necessarily, for any possible created person, it is within God’s power (whether or not) to choose that that person attain salvation.” Could a Dual Sourcer reject this last claim (the second conjunct of Predestination) on the grounds, for example, that God’s goodness necessitates that God choose the attaining of salvation for anyone he creates, in which case it would not be within God’s power to create a person for whom he does not choose the attainment of salvation? Such a move would not appear open to the Dual Sourcer, at least to the Dual Sourcer who affirms a libertarian reading of Responsibility. For, at least on the face of it, if it is incompatible with God’s essential goodness that he not choose salvation for every person he creates, then, given Responsibility, it is incompatible with God’s essential goodness that he not cause the accepting of the offer of salvation by every person he creates. But if God’s existence is incompatible with his not causing the acceptance of the offer of salvation for every person he creates, then God constitutes a prior and logically sufficient cause for any created person’s accepting the offer, in which case the person’s accepting is “determined” and not free in the libertarian sense. Although I do not want to rule out entirely the possibility of a Dual Sourcer rejecting the second conjunct of Predestination, in what follows I will assume that commitment

Providence, Grace, and Predestination

177

to the whole of Predestination is a part of Dual Sources and will turn to the question of whether that’s a good or a bad thing. One reason some might think it a bad thing is that Predestination has the following implication: that for any created person who does not attain salvation, God could have ensured that the person attain salvation by choosing that the person attain it. If, then, there are some who do not attain salvation, can a proponent of Predestination still say that God is just and good, that God loves all created persons, that God wills the salvation of all, and other things that many theists want to affirm? Let me close the book with several points in response to this question. First, Predestination and Dual Sources are compatible with almost any view regarding the ultimate landing place of human beings. While Predestination precludes the claim that God could not do otherwise than deliver the attainment of salvation for any persons he creates, it is perfectly consistent with the claim that God, in fact, delivers the attainment of salvation for every person. It is also consistent with the claim that very few attain salvation. Moreover, it is consistent with any view regarding the fate of those who do not make it. For instance, it is consistent with the claim that they are eventually annihilated, or with the claim that they spend eternity in Hell, conceived according to any model of Hell one likes, or with the claim that they live out eternity in some place other than Heaven or Hell. In short, Predestination and Dual Sources leave open the answers to these questions. In seeking answers, the Dual Sourcer must look to whatever he counts as evidence in such matters. Or a Dual Sourcer may remain agnostic. Second, it is not obvious that God’s not choosing the attainment of salvation for all persons he creates (even though he could do so) would conflict with God’s justice, goodness, love, or salvific will. Let us begin with God’s justice and goodness, prescinding at first from any appeal to scripture on the matter. I assume that justice does not require that God treat all people equally, that he give all equal gifts. I assume, rather, that to conflict with God’s justice, God would have to be wronging those he did not choose to attain salvation. And it might be plausible to think that God were wronging those he did not choose, if God owed all people the attainment of salvation, or if, even though God did not owe all people the attainment of salvation, God made it impossible, or rendered some people powerless, to attain it. But why think that God owes all people the attainment of salvation? Is it a natural right of human beings to be granted the beatific vision or eternal union with God? That would be, at the very least, a highly controversial claim. Certainly, God might owe the attainment of salvation to those who have done something to merit it. But those whom God does not choose, since they do not accept the offer of salvation, have not done and do not do what’s needed to merit it. God is depriving no one of her just reward. Moreover, on Dual Sources, as the argumentation in Section 4.5 should make clear, God’s not bringing about a person’s accepting the offer of salvation (and thus not choosing that they attain salvation) does not take away the person’s ability to have accepted the offer, all antecedent conditions remaining the same. Nor does it involve a positive act of reprobation, whereby God chooses that the creature not attain salvation in a way that “determines” the creature’s fate. Thus, God’s not choosing that I attain salvation would not have rendered me powerless to attain it. I could have accepted the offer and, had

178

Free Will and God’s Universal Causality: The Dual Sources Account

I done so, salvation would have been mine. If, then, I find myself on the outside of Heaven looking in, I might wish that God had given me the gift of salvation. But I will not have a claim that God has wronged me in any way. Even if we are satisfied that God’s not choosing salvation for all would not conflict with God’s justice, we might still think it would conflict with God’s goodness, unless we can articulate some reason why God would not have chosen the attainment of salvation for all, given that he could have. Considered in itself, a person’s salvation seems a good thing and the loss of salvation a bad thing. One might, then, think that a good God would necessarily choose salvation for all, unless there is some good to be secured only if not everyone attains salvation. Of course, skeptical theists have cautioned us about the inference from our not seeing a reason for God’s not doing something to the confident assertion that there is no reason. But, even apart from such skeptical considerations, reasons have been proposed for why God might not choose salvation for all. Echoing suggestions found in Augustine and Aquinas, Oliver Crisp, John Lamont, and Paul Macdonald have recently (and separately) proposed that God’s choosing salvation for some but not for others would enable a universe that most fully displays both God’s mercy and God’s justice: mercy to those who are saved and justice to those who suffer the consequences of their culpably not accepting God’s antecedent grace and offer of salvation.93 Thus, Crisp suggests the possible “need for the display of both God’s grace and mercy and his wrath and justice in his created order for some number of deserving humanity.”94 In the end, it is very difficult to judge whether God’s not choosing some for the attainment of salvation would be consistent with God’s justice, goodness, love, and salvific will, without making explicit our sources of evidence for understanding these latter divine attributes and what is and is not consistent with them. The two most common methods for arriving at an understanding of the divine attributes without appeal to divine revelation are first cause methodology and perfect being methodology. Here I simply confess that I find neither method very useful in answering the particular question at hand. First cause methodology, which makes inferences regarding what must be true or false of God if God is to explain whatever God is posited to explain (e.g., the existence or order of the universe), does not seem to give us a sufficiently detailed account of what is or is not consistent with God’s justice, goodness, love, or salvific will to answer the question at hand. I might, instead, try perfect being methodology, following my intuitions to arrive at a conception of the most perfect or worship-worthy being I can think of and of what that being would do vis-à-vis his creatures’ salvation. But it is not clear why I should think there is anything in reality corresponding to my conception. I am not, of course, denying that God is wholly perfect and worship worthy. My point is simply that, at least in my view, we cannot by these methods arrive at reliable expectations regarding God’s behavior on such a refined matter as whether or not God would grant salvation to every person he creates. In the end, I think many (most?) theists will arrive at their conception of God’s justice, goodness, love, and salvific will, with an eye on what they believe to be divine revelation. Certainly, this approach will be followed by most broadly traditional Christians. For this reason, at various points throughout the book, I have cited scriptural passages as a way of cautioning potential critics who might charge that Dual Sources

Providence, Grace, and Predestination

179

renders God too closely involved with sin. It seems that if one’s conception of God’s goodness is based in scripture, then one ought not confidently rule out as inconsistent with God’s goodness a level of involvement in sin, or a reason for permitting sin, of the sort that scripture appears, in multiple places, to attribute to God. Now, I fully admit that the proper interpretation of scripture is a difficult and controversial matter. But it is certainly not implausible to interpret scripture as taking up and giving an answer to the question of whether God’s not choosing that some attain salvation (when he could have) would be consistent with God’s justice and goodness. Arguably, St. Paul takes up this exact question (or something very similar) in Romans 9: What are we to say? Is there injustice on God’s part? By no means! … Has the potter no right over the clay, to make of the same lump one object for special use and another for ordinary use? What if God, desiring to show his wrath and to make known his power, has endured with much patience the objects of wrath that are made for destruction; and what if he has done so in order to make known the riches of his glory for the objects of mercy, which he has prepared beforehand for glory?”95

Whether Paul is saying here that some will, in fact, not attain salvation is not clear, but he certainly seems to be denying that it would be contrary to God’s justice were God not to save all. We have, then, scriptural reasons, as well as philosophical-theological reasons of the sort already mentioned, to deny that God’s justice and goodness necessitate, contrary to Predestination, that God must choose the attainment of salvation for every person he creates. Scripture proclaims God’s love for human beings and even God’s desire for the salvation of all.96 Are these scriptural teachings consistent with God’s not choosing that all attain salvation, when he could have? Well, either they are consistent or they are not. If they are, then these scriptural teachings must admit of a reading consistent with God’s not so choosing. Perhaps, God loves all human beings, but not in a sense that necessitates his delivering salvation for all; perhaps, he loves the elect in a different and greater sense than he loves the non-elect. Perhaps, when scripture speaks of God’s desire that everyone be saved, it refers to what God desires when considered in abstraction from goods that are possible only if some are not saved and for the sake of which God might not choose salvation for all. Or, perhaps, when scripture speaks of God’s desire that all be saved, it is referring to God’s giving all the offer of salvation and willing the antecedent grace a person needs to accept the offer. Such interpretations would be consistent with God’s not, in fact, bringing it about that all accept the offer, even though he could do so.97 On the other hand, if these scriptural teachings about God’s love and salvific will are not consistent with God’s not choosing that all attain salvation (when he could have), then it is perfectly consistent with Dual Sources and Predestination to hold that God does choose that all attain salvation and, thus, to hold with the universalist that everyone reaches Heaven. What Predestination (and Dual Sources, insofar as it entails Predestination) precludes is that God must choose the attainment of salvation for any

180

Free Will and God’s Universal Causality: The Dual Sources Account

person he creates. They rule out God’s being bound by nature or justice to choose the attainment of salvation for all. They do not preclude God’s actually choosing the attainment for all as a gift and act of love that goes beyond the requirements of nature or justice. In this connection, it seems that prayers for the attainment of salvation for ourselves and others make much more sense on the assumption that God is not bound by his justice or essential goodness to grant salvation to any person he creates. They make much more sense on the view that the attainment of salvation is God’s gift to give and something to hope for, not something that, in light of God’s essential goodness, God secures as a matter of necessity for all created persons. Of course, there are passages in scripture that appear strongly to suggest that not everyone attains salvation.98 Indeed, it might plausibly be thought that a. Scripture teaches Predestination, with its implication that the attainment of salvation is God’s gift, b. Scripture teaches that God desires salvation for all, and c. Scripture teaches that not all attain salvation. (a)–(c) are compatible only if one also holds d. Properly interpreted, what scripture means by God’s desiring salvation for all is consistent with God’s not choosing the attainment of salvation for all, even though he could have. If, however, one rejects (d) and assumes that scriptural teaching is consistent, then one must also reject one of (a)–(c). Thus, one guided by scripture in these matters faces the difficult interpretive question: Is the denial of one of (a)–(c) more plausible than affirming (d)? A Dual Sourcer, committed to Predestination, could answer “no.” Or she could answer “yes,” and, presumably, deny (c). Or she could think either (d) is true or, if not, then (c) is false, but not be sure which of these options is correct.99 Putting the question of scriptural interpretation aside, a potential advantage of denying Predestination is that if someone does not attain salvation, we need not hold that it was within God’s power to save the person, wondering why God didn’t. Indeed, a proponent of a view such as More Risky Open Theism, discussed in Chapter 7, could say that if some people do not attain salvation, God did absolutely everything within his power to promote their salvation; it is just that God’s power in this regard is limited. More generally, the weaker the account of divine providence, the easier it seems to distance God from any responsibility for the evils of the world—a result that many, understandably, find attractive. On the other hand, DUC and a strong account of providence have the attractive implication that all good—even that brought about by creatures—is, without remainder, directly given by God. Moreover, if one believes that God is good, just, and loving, it might be thought a more optimistic and hopeful view to affirm Predestination and a strong account of providence. For, on such a view, whatever evils exist are permitted by God for the sake of good purposes, whether known or unknown to us. And, since the attaining of salvation is in God’s hands, we can hope and pray that God delivers

Providence, Grace, and Predestination

181

salvation to ourselves and our loved ones, even those who appear far removed from accepting the offer of salvation, and who on a weaker conception of providence that denies Predestination, might appear outside God’s power to save. Dual Sources affords robust accounts of God’s providence, grace, and predestination, consistent with libertarian creaturely freedom, yet without the doctrine of middle knowledge. It reconciles libertarian freedom with divine universal causality, showing that libertarian theists need not reject a central tenet of the classic theological tradition. In performing these tasks, it does what many have assumed impossible. For this reason, Dual Sources deserves consideration alongside the alternatives for understanding God’s relationship to creaturely freedom.

Notes Chapter 1 1 Augustine, The City of God against the Pagans, 12.26, trans. R.W. Dyson (Cambridge: Cambridge University Press, 1998), 537. See also, Augustine, The Enchiridion on Faith, Hope, and Love, XI, trans. J.F. Shaw, in Basic Writings of Saint Augustine, Vol. 1, ed. Whitney J. Oates (New York, NY: Random House Publishers, 1948), 662: “Nothing exists but Himself [i.e., God] that does not derive its existence from Him.” 2 Anselm, Monologion 7, trans. Simon Harrison, in Anselm of Canterbury: The Major Works, ed. Brian Davies and G.R. Evans (Oxford: Oxford University Press, 1998), 20. Unless otherwise noted, all quotations from Anselm’s Monologion are taken from this source. 3 Maimonides continues: “All existing things, whether celestial, terrestrial, or belonging to an intermediate class, exist only through his true existence.” From Maimonides’s Mishneh Torah 1: 1–2. Translation from Isadore Twersky, ed., A Maimonides Reader (New York: Behrman House, Inc., 1972), 43. 4 Aquinas, Summa contra gentiles, 2.15.6, trans. James F. Anderson (Notre Dame: University of Notre Dame Press, 1975). Unless otherwise noted, all quotations from Summa contra gentiles, which I will henceforth abbreviate SCG, are taken from the Notre Dame Press edition. Note, however, that while James F. Anderson is the translator for Book II, Anton C. Pegis is the translator of Book I, Vernon J. Bourke is the translator of Book III, and Charles J. O’Neil is the translator of Book IV. 5 For extensive discussion of the scriptural support for God’s universal causality, see Section 2.1. 6 For discussion, see chapters 1–2 of G.L. Prestige, God in Patristic Thought (London: S.P.C.K., 1959); and Paul Copan and William Lane Craig, Creation out of Nothing (Grand Rapids, MI: Baker Academic, 2004), 124–45. 7 Indeed, Ludwig Ott claims that it is a de fide truth of the Catholic Faith that “all that exists outside God was, in its whole substance, produced out of nothing by God.” See Ludwig Ott, Fundamentals of Catholic Dogma (Charlotte: Tan Books, 2009), 79. 8 Alvin Plantinga, Does God Have a Nature? (Milwaukee: Marquette University Press, 1980), 2. 9 Brian Leftow, “Is God an Abstract Object?” Nous 24 (1990): 582. 10 Thomas V. Morris, Anselmian Explorations: Essays in Philosophical Theology (Notre Dame, IN: University of Notre Dame Press, 1987), 161. 11 Michael Bergmann and Jeffrey E. Brower, “A Theistic Argument against Platonism (and in Support of Truthmakers and Divine Simplicity),” Oxford Studies in Metaphysics 2 (2006): 361. 12 Hugh J. McCann, Creation and the Sovereignty of God (Bloomington, IN: Indiana University Press, 2012), 6.

Notes

183

13 William P. Alston, Divine Nature and Human Language (Ithaca, NY: Cornell University Press, 1989), 197. 14 Katherin Rogers, Anselm on Freedom (Oxford: Oxford University Press, 2008), 16. 15 Anselm, De concordia 1.7, trans. Thomas Bermingham, in Anselm of Canterbury: The Major Works, 447. Unless otherwise noted, all quotations from Anselm’s De concordia are taken from this translation. 16 Aquinas, Summa theologiae 1.44.2, trans. Fathers of the English Dominican Province (Westminster, MD: Christian Classics, 1981). Unless otherwise noted, all quotations from Summa theologiae, which I will henceforth abbreviate ST, will be taken from this translation. 17 ST 1.8.1. 18 Jonathan L. Kvanvig and Hugh J. McCann, “Divine Conservation and the Persistence of the World,” in Divine and Human Action: Essays in the Metaphysics of Theism, ed. Thomas V. Morris (Ithaca: Cornell University Press, 1988), 15. 19 Monologion 20, 34. My emphasis. 20 Mishneh Torah 1:2, 43. 21 SCG 2.21.3. My emphasis. The sentence immediately following points out that “of nothing else can this be said, for only He is the universal cause of being.” See also ST 1.8.4. 22 Does God Have a Nature?, 69. My emphasis. 23 ST 1.8.1. 24 ST 1.8.1 ad. 3. See also, SCG 3.68. 25 See especially Metaphysical Disputations 21.1 and 22.1. Unless otherwise indicated, all quotations from Suarez will be taken from Francisco Suarez, On Creation, Conservation, and Concurrence: Metaphysical Disputations 20–22, translation and introduction by Alfred J. Freddoso (South Bend, IN: St. Augustine’s Press, 2002). Henceforth, Suarez’s Metaphysical Disputations will be abbreviated as DM. 26 DM 21.3.3, 132. 27 Thomas V. Morris, Our Idea of God: An Introduction to Philosophical Theology (Notre Dame: University of Notre Dame Press, 1991), 154. 28 Contemporary action theorists use the term “basic action” to refer to an action that is done, but not by means of doing something else. See, for example, Jennifer Hornsby and Naomi Goulder, “Action,” in Routledge Encyclopedia of Philosophy, ed. E. Craig (London: Routledge, 1998, 2011). A “basic causal action” is just an application of this concept to causings or causal acts. If I break a window by throwing a baseball through it, my causing the window to break is not a basic causal action because I caused it to break by means of causing the ball to fly through it. Now, a couple of qualifications. First, although, on the traditional view, God directly causes all entities distinct from himself, there is a sense in which we can speak of creaturely causes as instrumental causes and of God as bringing certain effects about by means of other effects. The way in which this potentially misleading talk might be consistent with God’s directly causing all his effects is discussed in Section 3.2. Second, authors such as Aquinas did hold that God can be said to cause the operations of creatures in virtue of causing the powers by which the creaturely agents performed those operations (see, for example, SCG 3.67.2), in addition to causing those operations directly. Thus, strictly speaking, God would have been thought of as causing some entities both directly and indirectly. The key point is that no entities distinct from God are caused merely indirectly by him.

184

Notes

29 I use “entity” as a generic term covering positive ontological items of any sort, including substance, subject, accident, attribute, feature, trope, property, matter, form, essence, act of existence, state, action, etc., but not lacks or privations. By “distinct” I mean “not identical to.” I take “God” to refer to the same being or agent in all possible worlds or scenarios. I use “directly” and “immediately” as synonyms. I use “cause” and “bring about” interchangeably; thus, we could speak of God’s directly or immediately bringing about all entities distinct from himself. Notice that DUC would appear, also, to imply the doctrine of creation ex nihilo. Since no entities distinct from God exist that are not God’s effects, he does not bring about his effects out of entities that exist prior to or independently of his causing them. This point is consistent with its being the case that God’s causing one thing logically presupposes that he has caused some prior thing (i.e., God’s causing a virtue to come about in my soul presupposes that he has already brought about my soul). For more on how God’s action relates to time, see Section 8.5. 30 Whether it collapses into occasionalism will be the topic of Chapter 3. 31 Perhaps, a creaturely action is not a distinct entity over and above other entities that constitute it. But, if creaturely actions are, in this way, ontologically reducible to entities that constitute them, then they would still be caused by God, since the entities that constitute them would be caused by God. 32 Brian Davies, An Introduction to the Philosophy of Religion, second edition (Oxford: Oxford University Press, 1993), 42. My emphasis. 33 Anselm, On the Fall of the Devil 20, trans. Ralph McInerny, in Anselm of Canterbury: The Major Works, 223. Unless otherwise noted, all quotations from On the Fall of the Devil are taken from this translation. Recall that the claim that every action is caused by God is also affirmed by Anselm at De concordia 1.7, 447–48, a portion of which was quoted above: “Every quality, every action, everything that has existence owes its being at all to God” and “the very act of willing.… insofar as it exists, is something good and proceeds from God.” 34 Aquinas, Quaestiones disputatae de potentia dei 3.7. I will henceforth refer to this work as the De potentia and abbreviate it as DP. Unless otherwise noted, all quotations from the De potentia will come from On the Power of God, trans. the Fathers of the English Dominican Province (London: Burns, Oates & Washbourne Ltd, 1932). 35 Ibid. See also ST 1.105.5 and SCG 3.67. 36 ST 1.22.2 ad. 4. 37 DM 22.1.6, 152. Suarez cites a great number of scholastic and patristic authors supporting this view. Among the former group, he references Aquinas, Cajetan, Ferrariensis, Capreolus, Alexander of Hales, Albert the Great, Gregory of Rimini, Scotus, Bonaventure, and Hervaeus. Among the latter, he references Augustine, Cyril, Chrysostom, Nazianzen, Prosper, and Gregory the Great. See DM 22.1.6, 152–53. 38 ST 1–2.79.2. See also Aquinas, De malo 3.2. 39 DM 22.1.24, 163. My emphasis. That Anselm concurs with Aquinas and Suarez that even sinful acts, insofar as they are real, are caused by God can be seen at both On the Fall of the Devil 20 and De concordia 1.7. 40 Suarez, DM 22.1.25, 164. See also, DM 22.2.16 and 22.2.56. 41 SCG 3.70.8. 42 DM 22.1.22, 162. 43 For a helpful discussion of some of these details, with attention to both Suarez and his Dominican opponents, see Alfred Freddoso’s, “Introduction” in Suarez’s On

Notes

185

Creation, Conservation and Concurrence (South Bend, IN: St. Augustine’s Press, 2002), xcv–cxxi. I will, of course, have more to say about my own understanding of the relationship between God and creaturely agents in the pages that follow. 44 In Anselm on Freedom, Katherin Rogers also attempts to reconcile DUC (or what she calls the doctrine of God as creator omnium) with libertarian creaturely freedom. But, whereas I (and it would seem those I cite to this effect above) take DUC to imply that God causes our free choices and actions, Rogers denies this implication. For this reason, our projects are rather different. The burden of Rogers’s project is to show how one can embrace DUC while denying that God causes our free choices/acts. The burden of my project is to show how God can cause our free choices/acts, and yet they still be free in the libertarian sense. For appreciation and criticism of Rogers’s approach, see my “Anselm on Freedom: A Defense of Rogers’s Project, A Critique of Her Reconciliation of Libertarian Freedom with God the Creator Omnium,” The Saint Anselm Journal 8, no. 1 (2012): 1–10. Rogers offers a response in the same issue. Rogers takes up the case again in her Freedom and Self-Creation: Anselmian Libertarianism (Oxford: Oxford University Press, 2015). 45 To make a choice is to perform a certain kind of act and, so, in what follows, I will generally speak of “acts” or “actions” (terms I use interchangeably), rather than choices, allowing that choices can be understood as a type of action. For helpful introductions to contemporary freewill debates, see Timothy O’Connor, and Christopher Franklin, “Free Will,” in The Stanford Encyclopedia of Philosophy (Fall 2018 Edition), ed. Edward N. Zalta, https://plato.stanford.edu/archives/ fall2018/entries/freewill/; Robert Kane, A Contemporary Introduction to Free Will (Oxford: Oxford University Press, 2005); Kevin Timpe, Free Will: Sourcehood and Its Alternatives, second edition (London: Bloomsbury Academic, 2013); and Robert Kane, ed., The Oxford Handbook of Free Will (Oxford: Oxford University Press, 2002). 46 To avoid repetitive qualifications of this sort, I will assume in what follows that a view counts as “comprehensive determinism” if all events involving human beings are determined. Thus, “comprehensive determinism” would be compatible with indeterminism in some far-off region of the universe, or with God or angels performing acts that are not determined. 47 Kane, Contemporary Introduction, 5–6. 48 Anscombe, for example, seems prepared to endorse a meaning of “sufficient condition” other than the one employed in our definition of “determinism,” a meaning on which A could obtain, be a sufficient condition for B, and yet B not obtain. See G.E.M. Anscombe, “Causality and Determination,” in Metaphysics and the Philosophy of Mind: The Collected Papers of G.E.M. Anscombe, Vol. II (Minneapolis, MN: University of Minnesota Press, 1981), 135–37. On certain traditional accounts of grace, “sufficient” grace denotes grace the granting of which makes meritorious action possible without rendering it inevitable. On such accounts, then, “sufficient” clearly signals something other than what it signals when used to define determinism in contemporary discussions of free will. See Reginald Garrigou-Lagrange, Predestination (Rockford, IL: TAN Books and Publishers, 1998), 233–39. 49 Kane explicitly includes the adverb “logically” to modify “sufficient” in his definition of “determinism” at Robert Kane, The Significance of Free Will (Oxford: Oxford University Press, 1998), 8. Timothy O’Connor characterizes “determinism” as meaning that “the past (including my character and basic beliefs and desires) and the laws of nature logically entail that I do what I actually do.” See O’Connor, “Free Will,”

186

50 51

52 53 54

55

56 57

58 59 60 61 62 63

Notes section 3.2. According to Thomas Flint, “determinism” is the thesis that the universe at one time is a “logical consequence” of the universe at an earlier time together with the laws of nature. See Thomas P. Flint, “Compatibilism and the Argument from Unavoidability,” Journal of Philosophy 84 (1987): 424. Such references could be multiplied. See Timpe, Free Will, 22–23. Kane (see, for instance, Significance of Free Will) is a prominent incompatibilist who thinks we perform free actions. For a prominent incompatibilist who thinks we do not, see Derk Pereboom, Living without Free Will (Cambridge: Cambridge University Press, 2001). See, for instance, Kane, Contemporary Introduction, 32–33. For a view along these lines, see Timothy Pawl and Kevin Timpe, “Incompatibilism, Sin, and Free Will in Heaven,” Faith and Philosophy 26 (2009): esp. 405–13. In the case of our example, the agent is responsible for her character in virtue of the fact that it resulted from prior truthful acts or refrainings for which she is responsible. Although I will continue by speaking of “acts,” the following argument works just as well for “refrainings,” if these be viewed as other than acts. Even if one could perform an infinite number of acts between one’s very first act and some later act, and even if all but the first were acts for which one were derivatively responsible, the very first act would be one for which one was basically responsible and from which responsibility for the other acts would derive. Of course, between one’s very first act and any subsequent act, there is not time to perform an infinite number of acts, at least not conscious acts of the sort for which one could be morally responsible. Imagine someone going to the Sacrament of Reconciliation: “Bless me Father for I have sinned; it has been twenty minutes since my last Confession. Father, I have an infinite number of sins to confess since we last spoke.” I take “control,” “responsibility,” and “up to-ness” to be coextensive. Any event that is “up to” an agent is under the agent’s “control” and one for which the agent is “responsible,” etc. Readers may recognize the last three sentences as a version of the “Consequence Argument,” perhaps the most well-known argument for incompatibilism. A locus classicus for the Consequence Argument is Peter van Inwagen, An Essay on Free Will (Oxford: Oxford University Press, 1983), chapter 3. For a compact and accessible presentation of the argument, which has influenced my own presentation, see Thomas P. Flint, Divine Providence: The Molinist Account (Ithaca: Cornell University Press, 1998), 26–29. Alvin Plantinga, The Nature of Necessity (Oxford: Oxford University Press, 1974), 165–66. Ibid., 171. Roderick M. Chisholm, On Metaphysics (Minneapolis: University of Minnesota Press, 1989), 5–6. Ibid., 6–7. Eleonore Stump and Norman Kretzmann, “Eternity and God’s Knowledge: A Reply to Shanley,” American Catholic Philosophical Quarterly 72 (1998): 441, n. 8. Stump expresses a similar view in Aquinas (London: Routledge, 2003), 161. One infers from the passage that Stump and Kretzmann deny that Aquinas holds that God causes free human acts. As an interpretation of Aquinas, this denial is exceptional and, I believe, ultimately not sustainable in light of passages from Aquinas such as the ones quoted in Section 1.1.

Notes

187

64 Jeffrey E. Brower, “Simplicity and Aseity,” in The Oxford Handbook of Philosophical Theology, ed. Thomas P. Flint and Michael Rea (Oxford: Oxford University Press, 2009), 122. Brower’s agreement extends to the judgment that God’s causing our acts rules out libertarian freedom and not, as far as I know, to the interpretation of Aquinas regarding whether God causes our acts. 65 For other authors who suggest that God’s causing creaturely actions would commit us to a determinism incompatible with libertarian freedom, see Richard Swinburne, Providence and the Problem of Evil (Oxford: Clarendon Press, 1998), 33–35; Lynne Rudder Baker, “Why Christians Should Not Be Libertarians: An Augustinian Challenge,” Faith and Philosophy 20 (2003): 460–78; Edwin Curley, “The Incoherence of Christian Theism,” The Harvard Review of Philosophy 11 (2003): 84; and Anthony Kenny, Aquinas on Mind (London: Routledge, 1993), 77. 66 I speak of an agent’s “performing” its act, but I don’t take the performance to be anything distinct from the act itself. The performance is the act. By “voluntarily” here I mean that the act is something the agent does willingly, as opposed to being coerced. By “intentionally” I mean that the act is done on purpose. Some libertarians might hold that lack of determination, voluntariness, and intentionality are not sufficient for an act to be free in the libertarian sense. They might, for example, hold that these three do not suffice for the agential control necessary for free action. Perhaps, something further is needed, like an agent-causal relation. While I think this question ought to be taken seriously, I have here tried to define strict and broad accounts in a way that is as inclusive as possible of those philosophers who would call themselves, and be characterized as, libertarians. 67 A small minority of philosophers have, with varying degrees of explicitness, also challenged the prevailing view, at the very least denying that God’s causing a creaturely act entails that it is determined. See Hugh J. McCann, “Divine Sovereignty and the Freedom of the Will,” Faith and Philosophy 12 (1995): 582–98; McCann, Creation and the Sovereignty of God, esp. 103–12; Robert C. Koons, “Dual Agency: A Thomistic Account of Providence and Human Freedom,” Philosophia Christi 4 (2002): 397–410; Joseph Boyle, Jr., Germain Grisez, and Olaf Tollefsen, Free Choice: A Self-Referential Argument (Notre Dame: University of Notre Dame Press, 1976), 97–103; James F. Ross, “Creation,” The Journal of Philosophy 77 (1980): 615–19; and Brian Davies, The Reality of God and the Problem of Evil (New York: Continuum, 2006), 124–29. 68 The less specific term “neoscholastic” may be a more fitting characterization. First, although my approach is inspired by and draws from certain elements in Aquinas’s writings, important aspects of it are not found, or go beyond what can be found, in Aquinas. Second, given the speculative focus of the project, I make no attempt to show that every position I take is consistent with everything Aquinas says in his writings, and he admittedly says some things that appear to be (and maybe are) in tension with portions of my argument, even in cases where those very portions appear to be supported by other things Aquinas says. Third, there are aspects of Aquinas’s views on God and the created will that the present volume does not touch on, contrary to what might be expected of a “Thomistic” approach to these matters. Fourth, I do not engage the rich history of Thomistic interpretation in the way that might be expected of an approach characterizing itself as “Thomistic.” Finally, in light of the foregoing reasons, while I welcome discussion of the closeness (or lack thereof) to Aquinas, I do not wish my approach to be evaluated as an interpretation of Aquinas or as a presentation of his views; it is not intended as such.

188

Notes

69 ST 1–2.10.4. 70 William Hasker, “Providence and Evil: Three Theories,” Religious Studies 28 (1992): 97. For a more recent expression, which takes this point for granted, see Derk Pereboom, “Libertarianism and Theological Determinism,” Free Will & Theism: Connections, Contingencies, and Concerns, ed. Kevin Timpe and Daniel Speak (Oxford: Oxford University Press, 2016), 129. 71 I do not mean to imply that Molinism and Open Theism are the only popular alternatives for combining theism and libertarian freedom. For example, some libertarian theists reject Molinism but, in opposition to Open Theism, affirm that God knows future free creaturely acts by means of simple foreknowledge or in virtue of all of time being present to God’s eternity. Such approaches, however, do not typically include a distinctive account of how God’s sovereignty or providence (as opposed to foreknowledge) relates to human freedom. For this reason, such approaches are not as comprehensive as what Molinism, Open Theism, and Dual Sources attempt to offer.

Chapter 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

See also Rev. 4:11: “You are worthy, our Lord and God, to receive glory and honor and power, for you created all things, and by your will they existed and were created.” All quotations from the Bible are taken from the New Revised Standard Version. Cf. with reference to Christ, Col. 1:16–17. Walther Eichrodt, Theology of the Old Testament, Vol. II. trans. J.A. Baker (Philadelphia: The Westminster Press, 1967), 153–54. In the quoted text that follows, I replicate the notes that Eichrodt uses in referencing the relevant scriptural passages. Gen. 27:27; Ps. 65:7–14; Jer. 5:24, 10:13. Pss. 147:8ff; 145:15ff; Job 38:39ff. Gen. 26:12; 27:28; 49:25; Hos. 2:10ff; Ps. 107:35ff. Amos 4:6ff; Ps. 107:33ff. Gen. 15:5ff; 18:10ff; 25:21; 30:2, 8. Pss. 22:10ff; 90:3; 139:13ff; Job 10:8–12; 31:15; Isa. 41:4; Mal. 2:10. I Kings 17:17ff; II Kings 20:3; Ps. 6:2ff; cf. Ps. 107: 17ff. Ps. 107:23ff; Jon. 1:4. Isa. 9:7ff; Nah. 1:5; Job 9:6. Pss. 144:5; 104:32. Amos 5:8. Job 9:7, 9; Isa. 40:26; Ps. 147:4. Gen. 8:22. Job 9:10; Ps. 107:24. According to HarperCollins Bible Dictionary, “the heart.… represents the idea of the will, of human choice and conscience” [Entry by Douglas R. Edwards and Mark Allan Powell. Ed. Mark Allan Powell and others (New York: HarperOne, 2011), 368]. The Mercer Dictionary of the Bible states: “The association of the heart with reflection and intentionality leads to its recognition as the center of volition. One ‘wills’ with the heart (Jer. 23:20; Josh. 22:5; Luke 21:14; Col. 4:8). The heart, therefore, is the place for moral choice” [Entry by Scott Nash. Ed. Watson E. Mills and others (Macon, GA: Mercer University Press, 1990), 360]. The Interpreter’s Dictionary of the Bible states that the heart is “the center of the will and hence of the moral life” [Entry by

Notes

19 20

21 22

23 24 25 26

189

R.C. Dentan. Ed. George Arthur Buttrick and others (Nashville: The Abington Press, 1962), 550]. The Eerdmans Dictionary of the Bible states that “character, personality, will, and mind are modern terms which reflect something of the meaning of ‘heart’ in its biblical usage” [Entry by Larry L. Walker. Ed. David Noel Freedman and others (Grand Rapids: William B. Eerdmans Publishing Company, 2000), 563]. The King James Version reads: “Lord.… for thou also hast wrought all our works in us.” And the Revised Standard Version reads: “O Lord.… thou hast wrought for us all our works.” For consistency’s sake, this quotation comes from the NRSV, but apparently the verb “enabling” is not to be found in the Greek. A more accurate translation would be that of the RSV: “For God is at work in you, both to will and to work for his good pleasure.” My thanks to New Testament scholar John Martens, of the University of St. Thomas Theology Department, for help with this point. Cases where God appears not to know what his creatures will do in the future may also appear in tension with the claim that what they will do falls under his sovereign plan. For examples, see Jer. 26:3, Exod. 3:16–4:9, and Gen. 22:12. For other examples where God is causally behind bad actions, see Acts 4:27–28 (referenced above): “For in this city, in fact, both Herod and Pontius Pilate, with the Gentiles and the peoples of Israel, gathered together against your holy servant Jesus, whom you anointed, to do whatever your hand and your plan had predestined to take place.” See also, Gen. 50:20, where Joseph, hearing the apology of his brothers who, out of jealousy, had sold him into slavery, says: “Even though you intended to do harm to me, God intended it for good, in order to preserve a numerous people, as he is doing today.” For other examples of Jesus’s foreknowledge, see William Lane Craig, The Only Wise God (Grand Rapids: Baker Book House Company, 1987), 35–37. See, for example, Gen. 15:13–14; 1 Kgs 13:2; 2 Kgs 8:10–13; Dan. 2:39–43; Gen. 40:8; Deut. 31:16–17. Note that human choices are included within the content of many of these prophecies. See Isa. 41:21–23, 44.6–8, and 46.9–10. See also Deut. 18:22, which states: “If a prophet speaks in the name of the Lord but the thing does not take place or prove true, it is a word that the Lord has not spoken.” Those interested can explore these interpretive strategies by reading the work of Open Theists and their critics. Open Theists deny that God either causes or foreknows free creaturely actions, and they argue not just that their position is consistent with biblical teaching but that it is positively supported by scripture. These are difficult matters, but in my judgment there are too many scriptural passages that are not well accommodated by the Open Theists’ interpretive strategies. For a presentation and defense of the Open Theists’ strategies, see William Hasker, God, Time, and Knowledge (Ithaca: Cornell University Press, 1989), 190–96; Clark Pinnock, Richard Rice, John Sanders, William Hasker, and David Basinger, The Openness of God (Downers Grove: InterVarsity Press, 1994); John Sanders, The God Who Risks, second edition (Downers Grove: InterVarsity Press, 2007); and Gregory A. Boyd, God of the Possible (Grand Rapids: Baker Books, 2000). For criticisms of these strategies, see Craig, The Only Wise God, 21–44; Bruce A. Ware, God’s Lesser Glory: The Diminished God of Open Theism (Wheaton, IL: Crossway Books, 2000), 65–160; Bruce A. Ware, Their God Is Too Small (Wheaton, IL: Crossway Books, 2003), 25–57; and John M. Frame, No Other God: A Response to Open Theism (Phillipsburg, NJ: P&R Publishing, 2001).

190

Notes

27 For other examples, see Gen. 22:11–12 and Jer. 32:34–35. 28 Notice that even Open Theists must deny that the divine ignorance portrayed in the first passage (and the like) should be taken literally, for they are committed to God’s exhaustive knowledge of the present and past. Indeed, one of their main strategies for interpreting passages in which God predicts what free creatures will do is to say that even though God lacks certain knowledge of future free actions, God is able accurately to make many predictions about what free creatures will do based on his complete knowledge of the past and present, including full knowledge of the state of his free creatures’ characters. See, for example, Hasker, God, Time, and Knowledge, 194–95; David Basinger, “Practical Implications,” in The Openness of God, 163; Sanders, The God Who Risks, 133; and Boyd, God of the Possible, 35–37. Since Open Theists must take portrayals of God’s ignorance of the past and present nonliterally, they can hardly object in principle to a nonliteral reading of those passages that portray God as ignorant of the future. See Ware, God’s Lesser Glory, 65–86. 29 See Deut. 30:15–20, Ezek. 18, Sir, 15:14–17, Mt. 16:27, and Rom. 2:5–10. 30 In fact, I would maintain that from a pure reading of scripture, it is natural to affirm not just the negation of (4), but rather: (5) Scripture teaches that God causes human action and knows what humans will do. Together with (1) and (2), either (5) or the negation of (4) would entail the negation of (3). 31 Eichrodt, Theology of the Old Testament, 178. 32 Ibid., 179. 33 John M.G. Barclay, “Introduction,” in Divine and Human Agency in Paul and His Cultural Environment, ed. John M.G. Barclay and Simon J. Gathercole (New York: T& T Clark, 2006), 1. Barclay continues (1–2): “On the one hand, his [Paul’s] letters are full of statements which state or presuppose that human beings are capable and effective agents, responsible for their own actions. He bemoans sin as ‘disobedience’, and speaks of God’s judgment and wrath in ways that presuppose human guilt and human responsibility. He also cajoles, exhorts and instructs his converts as if they were both capable of, and responsible for, their own activity. On the other hand, he speaks as if God’s agency is effective everywhere, even in cases where humans are said to work. On the negative side, human sin is correlated with God ‘handing them over’ (Rom. 1.24, 26, 28); on the positive, human righteousness is identified with the leading of the Spirit (Gal. 5.18). Yet, even where all is of grace, human agency is not effaced; the dialectic of 1 Cor. 15.9–10 suggests that divine and human agency are not necessarily correlated to one another in inverse proportion.” 34 Anselm, Proslogion 2, trans. M.J. Charlesworth, in Anselm of Canterbury: The Major Works, ed. Brian Davies and G.R. Evans (Oxford University Press, 1998), 87. Unless otherwise noted, all quotations from Anselm’s Proslogion are taken from this source. 35 Proslogion 5, 89. 36 See ibid., 11 and 18. For a discussion of the approach, with attention also to Anselm’s Monologion, see Brian Leftow, “Anselm’s Perfect-being Theology,” in The Cambridge Companion to Anselm, ed. Brian Davies and Brian Leftow (Cambridge: Cambridge University Press, 2004), 132–56. 37 For examples, see Thomas V. Morris, both Our Idea of God and Anselmian Explorations (Notre Dame: University of Notre Dame Press, 1987); see also Katherin A. Rogers, Perfect Being Theology (Edinburgh: Edinburgh University Press, 2000). 38 Proslogion 5, 89. Emphasis added. 39 It would be a mistake to infer from the language “made everything else from nothing” that Anselm thinks that that than which nothing greater can be thought must be

Notes

191

said only to have made everything at the beginning of time. On the contrary, as is suggested by the title of the chapter, and as is stated explicitly elsewhere (see Monologion, Chapters 7 and 13), Anselm clearly holds that all that exists apart from God (whenever it exists) is made by God. 40 Morris, Our Idea of God, 40. Emphasis added. 41 Ibid. See also, p. 155, where Morris states: “Absolutely everything distinct from God depends on God for its existence. This is a foundational claim for any thoroughly theistic ontology. If God is the greatest possible being, a maximally perfect source of existence, then he is not just one more item in the inventory of reality. He is the hub of the wheel, the center and focus, the ultimate support, of all. The difference between theism and atheism is thus not just a disagreement over whether one entity of a certain description exists or not. It is a disagreement over the origin, and thus the ultimate nature, of everything.” 42 William Lane Craig, God Over All: Divine Aseity and the Challenge of Platonism (Oxford: Oxford University Press, 2016), 41. Craig here argues that a maximally perfect being would be the cause of anything distinct from itself in any possible world. And we could presumably derive the rest of DUC using perfect being methodology. For, it is plausible to think a being would be more perfect if things depended on it as on a direct or immediate cause and not simply an indirect, mediate, or remote cause. For yet another contemporary philosopher who uses perfect being methodology to support DUC, see Brian Leftow, God and Necessity (Oxford: Oxford University Press, 2012), 20–22 and 76. 43 Richard Swinburne, The Coherence of Theism (Oxford: Oxford University Press, 1993), 222. I do not agree with Swinburne, but I reference him to illustrate a difficulty to which perfect being methodology seems liable. 44 Morris (Our Idea of God, 35) characterizes the “core of perfect being theology” as the thesis that “God is a being with the greatest possible array of compossible greatmaking properties.” It follows from this thesis that if two great-making properties are not compossible, then God will have whichever of the two is greater, assuming that both are compossible with the greatest possible array of other great-making properties. 45 Katherin Rogers is a perfect being theologian who looks at points to accept such an argument against DUC. In Perfect Being Theology (101), she says, “If God produces your choice it isn’t free.” As mentioned in Section 1.1, Rogers, in fact, denies that DUC implies that God causes our free choices. 46 Typically, these arguments understand a “contingent” being to be a being that exists but might not have or that exists in the actual world, but not in some possible world. A “necessary” being, by contrast, is a being for which it is not possible that it not exist or a being that exists in all possible worlds. 47 Richard M. Gale and Alexander R. Pruss, “A New Cosmological Argument,” Religious Studies 35, no. 4 (1999): 462. 48 Ibid. I take it that to create the “universe,” understood as that which verifies or makes true our world’s Big Conjunctive Contingent Fact, means that God creates the particular entities that verify or make true the particular contingent propositions that make up that Big Fact. It would not make sense to speak of God’s creating what verifies or makes true that Big Fact, if many of the truths within that Big Fact were, in fact, not verified or made true by entities God created. 49 Robert C. Koons, “A New Look at the Cosmological Argument,” American Philosophical Quarterly 34, no. 2 (1997), 198–99.

192

Notes

50 Timothy O’Connor, Theism and Ultimate Explanation: The Necessary Shape of Contingency (Oxford: Blackwell Publishing, 2008), 83. 51 Ibid., 85. 52 I take it that the argument succeeds as an argument for monotheism only if reasons can be given for thinking it is a single necessary being that causes the aggregate of contingent entities. For such reasons, see Gale and Pruss, “A New Look at the Cosmological Argument,” 473; and O’Connor, Theism and Ultimate Explanation, 93. 53 Herbert McCabe, God Matters (Springfield, IL: Templegate Publishers, 1987), 11. 54 Germain Grisez, God? A Philosophical Preface to Faith (South Bend, St. Augustine’s Press, 2005), 275. 55 For Suarez’s arguments, see On Creation, Conservation, and Concurrence, 152–57. For discussion, see Freddoso’s “Introduction,” xcv–cxxi, as well as Alfred J. Freddoso, “God’s General Concurrence with Secondary Causes: Why Conservation Is Not Enough,” Philosophical Perspectives 5, Philosophy of Religion (1991): 553–85. 56 DM 22.1.13, 157. 57 DM 22.1.7, 154. 58 Peter van Inwagen, “The Place of Chance in a World Sustained by God,” in Divine and Human Action, ed. Thomas V. Morris (Ithaca, NY: Cornell University Press, 1988), 215–16, n. 4. 59 Plantinga, Does God Have a Nature?, 69. 60 Morris, Our Idea of God, 154–55. 61 Alston, Divine Nature and Human Language, 197. 62 Philip L. Quinn, “Divine Conservation, Secondary Causes, and Occasionalism,” in Divine and Human Action: Essays in the Metaphysics of Theism, ed. Thomas V. Morris (Ithaca: Cornell University Press, 1988), 54. Quinn (51) intends the variable x to range over concrete, contingent individuals. In explaining what he means by “brings about,” Quinn (52–53) says that the bringing about relation is characterized by “totality, exclusivity, activity, immediacy, and necessity.” Of particular interest to the present discussion is the characteristic of immediacy: “By immediacy, I mean that what does the bringing about causes what is brought about immediately rather than remotely through some causal chain or by means of instruments.” 63 I do not claim that these two are the only possible senses of “conserving” or “sustaining,” and my typology departs somewhat from the one given by Suarez at DM 21.3.2–3 (130–32) and discussed by Freddoso in his “Introduction” to Suarez’s On Creation, Conservation and Concurrence (lxxxviii–xci). Nevertheless, the two senses I distinguish will suffice for appreciating the force of Suarez’s argument. 64 Even on remote conservation x would cease to exist instantaneously, since the causal series in question is temporally synchronic, rather than diachronic. 65 Following Freddoso in “God’s General Concurrence with Secondary Causes,” 554. 66 Morris and Quinn make explicit that they are advocating immediate conservation, since they speak of the conserved entity’s direct or immediate dependence on God. The others are less clear on the matter, though I believe the most natural reading of their passages would take them to endorse immediate conservation as well. 67 In Suarez’s terminology, it is because they are both “beings through participation” that creaturely substances, and hence also creaturely operations, require God’s immediate causal influence: “If God does not have an immediate influence on every action of a creature, then the created action itself does not require God’s influence per se and essentially in order to exist, even though it, too, is a certain participation in being; … Created beings depend on God no less as agents than as beings, since

Notes

68 69 70 71

72

73

193

(i) they are no less subordinated to God for the one reason than for the other, and (ii) just as they are beings-through-participation, so too they are agents-throughparticipation; but insofar as they are beings, they are altogether dependent on God intrinsically and essentially; therefore, they depend on God in a similar way insofar as they are agents; therefore, while they are acting, they are dependent not only because they are being conserved in esse by God, but also because in their very acting they require God’s influence per se and immediately.” DM 22.1.9–10, 155–54. DM 21.1.9, 113. DM 22.1.9, 155. Suarez certainly knew of the argument as found in Aquinas and even references ST 1.8.1 and 1.104.1, where Aquinas gives versions of the argument. See DM 22.1.6–13, 111–15. Two remarks are in order in case one thinks I have misinterpreted Aquinas’s conclusion: First, although Aquinas doesn’t explicitly say in this text that God is the immediate, efficient cause of all being apart from himself, I think it safe to assume that Aquinas intends to be arguing for God as an efficient cause of all being and also as an immediate cause. Not only is that a natural reading of the text, but Aquinas gives a similar, if abbreviated, argument for the claim that created being is God’s proper effect at ST 1.8.1. This passage, quoted in Section 1.1, makes it clear that Aquinas thinks the cause of all being is an immediate, efficient cause. Second, because Aquinas says that “being” applies properly and unqualifiedly to substances, and not to accidents (see, for example, ST 1.90.2), it might be thought that Aquinas means to conclude only that God causes the being of all substances. Yet, the text says that “every being in anyway existing is from God,” which signals that Aquinas means all being, not just substantial being. Furthermore, in the article immediately following (ST 1.44.2), in arguing that prime matter is created by God, Aquinas states: “Whatever is the cause of things considered as beings, must be the cause of things, not only according as they are such by accidental forms, nor according as they are these by substantial forms, but also according to all that belongs to their being at all in any way.” I have not set out an ontology of creaturely operations, but that is not necessary to draw the present conclusion. Since whatever exists when a creaturely operation exists will be caused by God, it follows that the creaturely operation will be caused by God. At De malo, Question 3, Article 2, Aquinas actually gives a version of the Participation Argument in answering affirmatively the question whether God causes the act of sin; the passage provides additional evidence that Aquinas intends the Participation Argument to apply to all being, including the being of creaturely actions, and not just the being of substances. I realize that some may find strange the idea of a subject’s having a perfection to which it is identical, but abundant passages make clear that Aquinas thinks to have a perfection essentially is for a subject to be identical to that perfection. See, for example, his Commentary on Aristotle’s Metaphysics, 1.10.154; his Quodlibetal Questions 2.2.1; his Commentary on the Book of Causes 9; and his Exposition on the “On the Hebdomads” of Boethius 2. In the passage in which Aquinas presents the Participation Argument (ST 1.44.1), we find the claim that “all beings apart from God are not their own being, but are beings by participation.” The locution “not their own being” (non sunt suum esse) signals a lack of identity between the subject to which being belongs and its being, which suggests that to have being (or any other perfection) by participation is to have being without being identical to it, whereas

194

Notes

to have being (or any other perfection) essentially is to be identical to being, as God is. See, further, ST 1.3.1, ST 1.44.3, and ST 1.45.5 ad 1. For helpful discussion, see W. Norris Clarke, “The Meaning of Participation in St. Thomas,” Proceedings of the American Catholic Philosophical Association 26 (1952): 147–59; and Jan Aertsen, Nature and Creature: Thomas Aquinas’s Way of Thought (Leiden: E.J. Brill, 1988), 79–91. 74 Rudi te Velde discusses the premise during a helpful treatment of Aquinas’s Participation Argument in Aquinas on God: The “Divine Science” of the Summa Theologiae (Burlington, VT: Ashgate, 2006), 128–32. He does not, however, offer a defense of the premise. 75 See Quodlibetal Questions 2.2.1; SCG 2.53.4 and 2.54; and ST 1.75.5 ad 1 and 4. For discussion, see W. Norris Clarke, The One and the Many: A Contemporary Thomistic Metaphysics (Notre Dame: University of Notre Dame Press, 2001), 156–58; and W. Norris Clarke, Explorations in Metaphysics (Notre Dame: University of Notre Dame Press, 1994), 65–88. 76 To say that the perfection as such enters into, and is explanatorily prior to, the composition through which a subject has that perfection by participation does not imply that the perfection as such has any sort of existence outside of the compositions into which it enters. In a similar way, Aquinas thinks of prime matter as a principle or metaphysical component of material substances, and thus as explanatorily prior to those substances, without thereby thinking that prime matter ever exists independently of material substances. 77 ST 1.4.2. 78 I note also that were it necessary for a caused perfection to be found in its cause according to the exact same formality, then Aquinas’s acceptance of premise (1) would commit him to the existence of, for example, subsistent human natures, something which Aquinas explicitly denies. For, Aquinas thinks that natures are perfections belonging by participation to the individuals that have them. Thus, if whatever perfection belongs to something by participation must be caused in it by that to which it belongs not only essentially but also according the same species or formality, then there would be at least one human nature (and, similarly, one canine nature, one feline nature, etc.) that was its own subsisting subject. Aquinas denies this implication of premise (1) by claiming that, when it comes to the natures of material things, the individuals that have those natures by participation have those natures caused in them by agents to which the perfection of those natures belongs essentially, but according to a higher or more eminent formality, not according to the very same formality. For the foregoing, see ST 1.44.3, especially objection 2 and the response to objection 2. 79 I have offered evidence that Aquinas would accept the chief claims from which our defense of premise (1) has been constructed. That he would accept the defense is also suggested by certain passages in which he argues that individuals participating in a particular form or nature cannot cause that form as such in other individuals of the same nature, and precisely because all such individuals presuppose that form to exist as the kinds of things they are. See SCG 2.21.5, SCG 3.65.4, DP 5.1, ST 1.104.1, and ST 1.45.5 ad 1. 80 How many subsisting perfections are there? As far as I know, Aquinas does not tell us. But the metaphysical claims on which the defense of premise (1) has been constructed require only that there be one subsistent perfection, God, which contains the perfection of everything else.

Notes

195

81 This point is clear, for example, from the passages cited in the note before last. 82 See Barry Miller, The Fullness of Being: A New Paradigm for Existence (Notre Dame: University of Notre Dame Press, 2002); Barry Miller, A Most Unlikely God: A Philosophical Enquiry (Notre Dame: University of Notre Dame Press, 1996); David S. Oderberg, Real Essentialism (New York: Routledge, 2007), 121–30; William F. Vallicella, A Paradigm Theory of Existence: Onto-Theology Vindicated (Dordrecht: Kluwer Academic Publishers, 2010); and Edward Feser, Scholastic Metaphysics: A Contemporary Introduction (Heusenstamm: Editiones Scholasticae, 2014), 241–56. 83 See ST 1.4.1 and 1.5.1. 84 See, for instance, De ente et essentia 4. 85 See De ente et essentia, 4–5; ST 1.50.4; and SCG 1.42. It is unclear whether this argument that for any perfection P, there can be only one subsistent P rules out there being a subsistent P and also a subsistent P*, where P* is a higher perfection containing the perfection of P, but according to a higher formality. Fortunately, we can avoid answering this question in considering the Participation Argument. For, Aquinas thinks there is no higher perfection than that of subsistent existence itself, which contains within itself all perfection (see SCG 1.28.2 and ST 1.4.2). Given this, it is not possible that there be both subsistent existence and another subsistent perfection, higher than subsistent existence, but containing the perfection of subsistent existence.

Chapter 3 For a classical critic of occasionalism, see Aquinas, SCG 2.69. For some recent critics, see Katherin A. Rogers, “What’s Wrong with Occasionalism?” American Catholic Philosophical Quarterly 75, no. 3 (2001): 345–69; van Inwagen, “The Place of Chance in a World Sustained by God,” 215–16, n. 4; and Quinn, “Divine Conservation, Secondary Causes, and Occasionalism,” 55 and 73. Although McCann criticizes occasionalism in Creation and the Sovereignty of God (35–38), as we will see, he also denies the compatibility of DUC and productive creaturely causation. 2 Suarez, Disputationes metaphysicae 22.1.2, 162. For Aquinas, see Summa contra gentiles 3.70.8. 3 See Nicholas Malebranche, Dialogues on Metaphysics and on Religion, ed. Nicholas Jolley and David Scott (New York: Cambridge University Press, 1997), 106–11. 4 Timothy D. Miller, “Continuous Creation and Secondary Causation: The Threat of Occasionalism,” Religious Studies 47 (2011): 8. 5 Quinn, “Divine Conservation, Secondary Causes, and Occasionalism,” 71. 6 William Vallicella, “Concurrentism or Occasionalism?” American Catholic Philosophical Quarterly 70, no. 3 (1996): 349. 7 Hugh J. McCann and Jonathan L. Kvanvig, “The Occasionalist Proselytizer: A Modified Catechism,” Philosophical Perspectives 5, Philosophy of Religion (1991): 590. 8 Ibid., 613, n. 14. 9 See Vallicella, “Concurrentism or Occasionalism,” 351–59; and ibid., 610–13. In his more recent Creation and the Sovereignty of God (38–42), McCann offers an account of secondary causation that understands secondary causation to involve not the production of an effect by its cause but rather the transfer of an already existing quantity (e.g., energy) from cause to effect. On this account, a secondary cause 1

196

10

11

12

13 14

15

16 17

18

Notes does not bring its effect into existence, for the effect is just a new manifestation or development or state of what was already there. In their discussions of productive (or existence-conferring) secondary causation, Vallicella focuses on causation by substances while McCann and Kvanvig focus on causation by events. Both Vallicella and McCann/Kvanvig offer criticisms of the kind of causation they focus on that are independent of their belief that God is the productive cause of all that exists. Although they are worthy of consideration, these independent arguments will not concern us here. I limit myself to considering whether these authors are correct in their claim that affirming an effect to have both God and a creature as productive causes would render one or the other of these causes otiose. Proponents of NODUC will acknowledge that some of God’s effects do not also have creaturely causes—for example, what God brings about at the first instant of time, as well as certain miracles brought about by God, but not also by any secondary cause. A similar distinction between “metaphysical” and “epistemic” objections to causal overdetermination can be found at Theodore Sider, “What’s So Bad about Overdetermination?” Philosophy and Phenomenological Research 67, no. 3 (2003): 719–25. This section presupposes and builds on my introduction of DUC in Section 1.1. Yet, my responses to the metaphysical and epistemic objections below would, I think, work just as well were efficient causes identified as events. For helpful discussions of efficient causality in the scholastic tradition, see Freddoso, “Introduction,” xliii–lxxiii; Alfred J. Freddoso, “God’s General Concurrence with Secondary Causes: Pitfalls and Prospects,” American Catholic Philosophical Quarterly 67, no. 2 (1994): 135–42; Stephen L. Brock, Action and Conduct: Thomas Aquinas and the Theory of Action (Edinburgh: T&T Clark, 1998); Michael Rota, “Causation,” in The Oxford Handbook of Aquinas, ed. Brian Davies and Eleonore Stump (Oxford: Oxford University Press, 2012), 104–14; and Feser, Scholastic Metaphysics, ch. 2. For contemporary metaphysicians, in addition to the foregoing, who take efficient causes to be substances, see E.J. Lowe, Personal Agency (Oxford: Oxford University Press, 2008), 121–78; Oderberg, Real Essentialism, 130–43; and William Hasker, The Emergent Self (Ithaca: Cornell University Press, 1999), 100–01. It may be argued, for various reasons, that fire is not a genuine substance, and therefore that the fire’s causing the heat is not a genuine example of substance causation. For the convenience of the example, we will simply assume (or pretend) that the fire is a substance and invite the reader to substitute another example, if they find it more helpful. See, for instance, DP 3.7 corpus and ad 16; SCG 2.21.5 and ST 1–2.6.1 ad 3. In what does the fire’s causal act or producing the heat consist? Possible answers are that the fire’s causal act does not consist in any additional entity distinct from the heat, or the heat and its relation of dependence on the fire, or the heat, the dependence, and the fire together. For one who holds one of these views, God’s concurrence in causing the fire’s causal act will not be anything over and above his concurrence in causing whatever items make up the act. Whether God’s act or operation and the creaturely agent’s act or operation are identically one and the same act is a question I will briefly take up in Chapter 5. Although the arguments of this chapter do not turn on that question, we will typically speak as if God’s operation and the creature’s operation are distinct.

Notes

197

19 Freddoso, “Introduction,” xcviii. In addition to “Introduction,” xcvi–c, Freddoso helpfully discusses the following in “Pitfalls and Prospects,” 131–56. 20 Freddoso, “Pitfalls and Prospects,” 144. 21 Freddoso, “Introduction,” xcviii. Emphasis added. 22 See Freddoso, “Introduction,” xcviii–xcix, and “Pitfalls and Prospects,” 144–51. 23 Freddoso, “Introduction,” xcviii. 24 See Aquinas, SCG 3.70.8; and Suarez, DM 22.1.22. These passages were quoted in Section 1.1. 25 ST 1.44.2. 26 For an interesting attempt to reconcile creaturely freedom with God’s universal causality by tracing the being of the act to God, but the essence or form to the creature, see Mark K. Spencer, “Divine Causality and Created Freedom: A Thomistic Personalist View,” Nova et Vetera 14, no. 3 (2016): 375–419. My question, as above, is whether this sort of approach really preserves DUC. 27 Unlike me, Freddoso (“Pitfalls and Prospects,” 150, n. 27) is “inclined to think [that] any coherent version of concurrentism will acknowledge that God and secondary agents are primarily responsible for different facts or states of affairs involving the termini of causal relations,” and his reason seems to be that, otherwise, their contributions would be “superfluous” or “redundant” (see 145 and 147). The strongest evidence Freddoso gives that the view he describes is the traditional scholastic view are some passages from Aquinas (“Pitfalls and Prospects,” 146, ST 1.105.5; SCG 3.66.6; DP 3.1). While Freddoso’s reading of these passages is certainly plausible, it is not, perhaps, necessary to read the passages as suggesting that the determinate character or form of the effect is not wholly from God as well as from the creaturely agent, or suggesting that there is some aspect of the creature’s effect that is not really the creature’s effect. 28 It is possible that knowing the creature’s activity, together with other premises, would enable one to know by means of an inference God’s activity in causing the effect. But to know the creature’s activity would not, by itself, be enough to know God’s activity. 29 Some authors speak of God and creaturely causes as operating on different and non-competing causal levels or plains. In terms of the foregoing, we might say that God and creaturely causes operate on different levels, since the creature, its effect, and its bringing about of its effect are all included within the object that God brings about, but not vice versa. At the same time, God and creatures are not competing causes, as if the agency of one were inversely proportional to the agency of the other. On the contrary, the effect shared in common is wholly caused by both. Indeed, the secondary cause’s producing the whole of the effect is part of what is brought about in the higher level divine causal activity. 30 In the remainder of this chapter, I will for the sake of brevity sometimes speak of “the heat’s being caused or brought about by both God and the fire” as shorthand for “the heat’s being brought about by God, and the heat’s also being brought about by the fire.” As I will be using it, the shorter locution “being brought about by both” should not be read as claiming that God and the fire “cooperate” in bringing about the heat, where that means each only brings about a part of the heat, and not the whole. As discussed in the previous section, according to NODUC, God and the fire “cooperate”; they don’t “cooperate.” 31 Sider, “What’s So Bad about Overdetermination?” 721. 32 McCann and Kvanvig, “Occasionalist Proselytizer,” 590. Emphasis added. Sider says of the foregoing that it is a “bad picture,” which “takes seriously a view of

198

Notes

causation that no one accepts.” See Sider, “What’s So Bad about Overdetermination?” 721. Whether McCann and Kvanvig accept something like this picture cannot be determined from their brief statement of the worry. 33 Miller, “Continuous Creation and Secondary Causation,” 8. 34 Showing that SC1 and SC2 are incompatible would show not that it is impossible for a single effect to be brought about by two distinct causes, but only that it is impossible that a single effect be brought about by one cause (the secondary cause) if that cause cannot bring about the effect unless the effect is also brought about by another cause (God). 35 The three ways Miller considers all admit of nontheological examples, where one creaturely agent receives assistance from another. As noted in the previous section, however, it is likely (certainly possible) that the relationship between God and creaturely causes, and thus the kind of concurrence/assistance creatures require from God, is sui generis. Miller never denies this possibility. 36 Miller, “Continuous Creation and Secondary Causation,” 8. 37 Ibid., 9. 38 Notice a secondary cause’s having all the causal power needed to bring about an effect is not mentioned in Miller’s statement of SC1. 39 I take it that x is causing y at a time t if y causally depends on x at that time. 40 Alston, Divine Nature and Human Language, 202–03. 41 And, given NODUC, in one world, but not the other, God brings about the fire’s causing the heat. 42 As Aquinas says at De potentia 3.7 ad 16: “God can produce the natural effect even without nature: but he wishes to act by means of nature in order to preserve the order in things.” 43 Agreement with this last judgment would be required by proponents of a commonly accepted view that there can be cases of causal overdetermination, where a single effect has two causes, either of which would have caused the effect without the other; for instance, a victim is shot simultaneously by two assassins, either of whose bullets is itself sufficient for the victim’s death. For examples of philosophers who appear to recognize that there can be such cases, see Jaegwon Kim, “Mechanism, Purpose, and Explanatory Exclusion,” Philosophical Perspectives 3, Philosophy of Mind and Action Theory (1989): 91–92; Karen Bennett, “Why the Exclusion Problem Seems Intractable, and How, Just Maybe, to Tract It,” Nous 37, no. 3 (2003): 477; and Brandon Carey, “Overdetermination and the Exclusion Problem,” Australasian Journal of Philosophy 89, no. 2 (2011): 252–53; and Thomas D. Bontly, “Exclusion, Overdetermination, and the Nature of Causation,” Journal of Philosophical Research 30 (2005): 263. 44 For examples of philosophers who take our experience of causation to give us reason to believe that things in the world, including ourselves, are efficacious, see Anscombe, “Causality and Determination,” 136–38; John Searle, Mind: A Brief Introduction (Oxford: Oxford University Press, 2004), 142–44; and Hasker, The Emergent Self, 107–09. 45 Jaegwon Kim, Physicalism, or Something Near Enough (Princeton: Princeton University Press, 2005), 42. See pages 32–69 for the use to which Kim puts the principle. 46 Kim, “Mechanism, Purpose, and Explanatory Exclusion,” 93. 47 Ibid. 48 Ibid.

Notes

199

49 Ibid., 90. Emphasis added. A page earlier in the essay (89), Kim defines the principle as follows: “The general principle of explanatory exclusion states that two or more complete and independent explanations of the same event or phenomenon cannot coexist.” The wording of the exclusion principle has obviously changed from this 1989 essay to the statement of it from 2005, quoted above. Although Kim does not explain precisely how he is using the term “sufficient cause” in the 2005 statement (or in other recent statements I have checked), I assume that Kim believes that if an event has “more than one sufficient cause occurring at any given time,” then these causes are not complete and independent, unless it is a genuine case of causal overdetermination. 50 Kim, “Mechanism, Purpose, and Explanatory Exclusion,’’ 90–91. 51 This present point notwithstanding, the claim made two paragraphs above—that assuming two causes of a single effect possible, it can be reasonable to affirm both, if the evidence for both is strong—holds good even for causes that are complete and independent of one another. Indeed, Kim himself acknowledges that there are cases of causal overdetermination, where a single effect has two causes that are independent of one another and either of which would have brought about the effect in the absence of the other. See Kim, “Mechanism, Purpose, and Explanatory Exclusion,” 91; and Kim, Physicalism, or Something Near Enough, 42. 52 Summa contra gentiles 3.69. 53 See Peter van Inwagen, “God and Other Uncreated Things,” in God and Metaphysics: Essays in Honor of Eleonore Stump, ed. Kevin Timpe (New York: Routledge, 2009), 3–20. In this essay, van Inwagen allows that there are certain entities, namely, abstract objects, which God does not bring about. For this reason, he would not count as a proponent of DUC, given the way I have defined it and the way it has generally been understood by the theistic tradition, as evidenced in Section 1.1. van Inwagen still wants to affirm the traditional Nicene creedal statement that God is maker of all things visible and invisible. He interprets the quantifier “all things” in such statements to be restricted to things that can enter into causal relationships and thus as leaving abstract objects outside the class of entities that God is professed to make. 54 van Inwagen, “The Place of Chance in a World Sustained by God,” 216. 55 McCann and Kvanvig, “Occasionalist Proselytizer,” 614. 56 The question is paraphrased from Vallicella, “Concurrentism or Occasionalism?” 349, and Miller, “Continuous Creation and Secondary Causation,” 4. 57 For an excellent survey of some of these accounts, see The Oxford Handbook of Free Will, 337–437. 58 See Hasker, The Emergent Self, 107–09. 59 Timothy O’Connor, Persons and Causes (Oxford: Oxford University Press, 2000), 52–61. 60 Lowe, Personal Agency, 129. 61 Perhaps, O’Connor would welcome this conclusion; for, as we saw in Section 2.3, in more recent work he offers an argument for God’s existence in which he contends that “the existence of each natural particular and the events in which they participate admit, in principle, of a fully adequate explanation in terms ultimately involving their causal dependency on a necessary being.” Surely, agentcausal acts would be included among “the existence of each natural particular and the events in which they participate.” See O’Connor, Theism and Ultimate Explanation, 85.

200

Notes

62 Richard Taylor, Action and Purpose (Englewood Cliffs, NJ: Prentice-Hall, Inc., 1966), 111–16. 63 O’Connor, Persons and Causes, 52. 64 Ibid., 52–53. O’Connor does not explain what he means by “metaphysically basic instance of causation,” but I assume he does not mean “an instance of causation that has no cause”; for, if that were his meaning, his claim that metaphysically basic instances of causation have no cause would be a tautology, and, instead of arguing for this claim (as he does), he would need to argue that agent-causal actions are, in fact, metaphysically basic instances of causation. 65 O’Connor, Persons and Causes, 53. 66 Ibid. 67 Ibid. 68 Ibid. 69 Ibid. 70 In an earlier essay [“Agent Causation,” in Agents, Causes, and Events ed. Timothy O’Connor (Oxford: Oxford University Press, 1995), 186], O’Connor focuses on the first sort of example, allowing that complex events of the form event X’s causing event Y can be caused, but only in a derivative way, by causing event X. While denying that this sort of example has application to agent-causation, he also states: “There cannot be an immediate, efficient cause of a causal relation (i.e., independently of the causation of its front end relatum).” If this claim were true, then it would, given a certain ontology of relations, rule out NODUC’s claim that creaturely causings (like all else apart from God) are immediately, efficiently caused by God. O’Connor, however, does not offer support for the claim. 71 O’Connor, Persons and Causes, 55–60. The reader can consult these pages for various works of Chisholm referenced by O’Connor. 72 O’Connor, Persons and Causes, 61. 73 Ibid., 58–59. 74 Ibid., 58. 75 Ibid., 61. 76 Claim (iii) is the meaning that O’Connor gives to “deficient” in the text.

Chapter 4 1

2 3 4 5 6

For a discussion of the distinction, see Dan Marshall and Brian Weatherson, “Intrinsic vs. Extrinsic Properties,” in The Stanford Encyclopedia of Philosophy (Spring 2018 Edition), ed. E.N. Zalta, http://plato.stanford.edu/archives/spr2018/entries/ intrinsic-extrinsic/. See Brower, “Simplicity and Aseity,” 124, n. 1. Of course, it is grounded in things inside Elizabeth, too, namely, her belief. But it is also grounded in extrinsic items, and, hence, the predication is extrinsic. The term was introduced by Peter Geach, God and the Soul (New York: Schocken Books Inc., 1969), 71–72. As in my definition of DUC, I use “entity” as a generic term covering positive ontological items of any sort, including substance, accident, matter, form, property, feature, state, event, action, operation, etc., but not lacks or privations. For other authors who define the intrinsic–extrinsic distinction in terms of the distinction between real and merely Cambridge change, see Peter van Inwagen,

Notes

7

8

9

10

11

12

13 14

201

Metaphysics (Boulder: Westview Press, 1993), 33–34; Harold Noonan, Personal Identity (New York: Routledge, 1989), 162–63; E.J. Lowe, The Possibility of Metaphysics (Oxford: Clarendon Press, 1998), 93. Instead of “extrinsic” properties, van Inwagen speaks of “relational” properties and Noonan speaks of “mere Cambridge” properties, but the meaning is the same as my “extrinsic.” As I noted when giving these definitions in Section 1.2, some libertarians might hold that voluntariness and intentionality, along with lack of determination, is not sufficient for an act to be free. They might, for example, hold that these three do not suffice for the agential control necessary for free action. Perhaps, something further is needed, like an agent-causal relation. While I think this question ought to be taken seriously, I have here tried to define the strict and broad accounts in a way that is as inclusive as possible of those philosophers who would call themselves, and generally be characterized as, libertarians. Philosophers of religion frequently write in ways that suggest that it is not just God but God’s causal or creative activity, which accounts for God’s effects and on which those effects depend or from which they derive. Indeed, though we have not yet drawn attention to it, we find this suggestion in many passages already quoted in this volume. See, for example, the passages from Brower and Bergmann, Plantinga, and McCann in Section 1.1. For example, see Quinn, who in “Divine Conservation, Secondary Causes, and Occasionalism” writes (54): “Necessarily, for all x and t, if x exists at t, God willing that x exists at t brings about x existing at t.” Emphasis added. See also, Alston, Divine Nature and Human Language, 154: “Since God is bodiless He acts in the world by directly producing worldly effects of His volitions, not by producing those effects by movements of His body.” Emphasis added. I do not take the decree or choice’s causing E and its being that in virtue of which God causes E to be mutually exclusive. Indeed, it would seem that any proponent of the view who held that God’s choice causes E would also need to hold that the choice is that in virtue of which God causes E, since he will affirm that God causes E, and so presumably need to say that God does so by means of his choice or decree. In what follows, I will therefore simply speak of the choice or decree as that in virtue of which God causes E. The name “extrinsic model” is my own. For recent defenders, see Timothy O’Connor, “Simplicity and Creation,” Faith and Philosophy 16, no. 3 (1999): 405–12; Brower, “Simplicity and Aseity”; Alexander R. Pruss, “On Two Problems of Divine Simplicity,” Oxford Studies in the Philosophy of Religion 1 (2008): 150–67; Miller, A Most Unlikely God, esp. 106–13; and my “Must a Cause Be Really Related to Its Effect? The Analogy between Divine and Libertarian Agent Causality,” Religious Studies 43 (2007): 1–23. For a classic exposition of the doctrine, see Aquinas, ST 1.3. One reason scholastic proponents of divine simplicity deny that there are entities (such as accidental properties) intrinsic to God but distinct from God is that they understand such entities to enter into the metaphysical composition of the thing that has them; and, of course, divine simplicity rules out any composition in God. See Brian Leftow, “Divine Simplicity,” in Routledge Encyclopedia of Philosophy, ed. E. Craig (London: Routledge, 1998). For Aquinas on God’s freedom to do otherwise, see ST 1.19.3–4; ST 1.25.5 corpus and ad 1&3; ST 1.25.6 corpus and ad 6; SCG 2.23; and DP 3.15. In Chapter 5, I will consider a way in which one committed to divine simplicity and God’s freedom to do otherwise might nevertheless still attempt to identify God’s

202

15 16

17 18 19 20 21

22

23

24 25

26

Notes causing E, or choosing to cause E, with God. As we will see, this way introduces certain drawbacks that are avoided by EM. Mark G. Henniger, Relations: Medieval Theories 1250–1325 (Oxford: Oxford University Press, 1989), 7. The claim that there is no real foundation in God for the relation does not mean that there is nothing intrinsic to God at all that the relation presupposes. For example, a causal relation holding between God and some effect certainly presupposes that God has the power to bring about that effect. But, as we will see, the claim does imply that there is nothing intrinsic to God that would not be there were God not so related. To avoid unnecessary complications in what is merely an attempt to explain the original context for EM, I will forego further discussion of scholastic accounts of foundations. See, for instance, Aquinas at Summa contra gentiles, Book II, Chapters 11 and 12. O’Connor, “Simplicity and Creation,” 408–09. Were God not causing E, E would not exist given the assumption that E can’t exist unless caused by God. This assumption would, of course, be endorsed by proponents of DUC. On EM, God’s reasons for causing what he does are not entities distinct from God. Rather, they are contained in God’s knowledge of himself, much in the same way that Aquinas says the divine ideas are. See Aquinas, Summa theologiae 1.15.1–2. Strictly speaking, EM and the fundamental arguments of this book would be consistent with saying that God sometimes has reasons that, given his wisdom and goodness, do not leave him free from bringing about the effect in question. In order to avoid increased complications and qualifications in what will already be a demanding discussion, I will proceed (unless otherwise stated) assuming a version of EM on which God’s reasons always leave him free from bringing about the effect in question. Again, I take God’s “causing” and God’s “causal act” to refer to the same thing. In Chapters 7 and 8, I will say more about God’s willing. The limited, but important, point of the foregoing is to make clear that God can intentionally bring about his effects, and can will and choose them, without this implying that these are intrinsic states of God. In Section 5.3, I will have more to say about various ways of understanding the ontology of the causal relation. In that context, a third possibility will emerge for understanding what God’s causal act consists in, on which God’s causal act is simply identified with the effect E. And this whole may be brought about for the sake of some other end, such as God’s glory or goodness. Under what conditions an agent’s reason can unite into one action various things the agent brings about is a very difficult and controversial question in action-theory, which I oversimplify here for the sake of brevity. For present purposes, it is enough to recognize, in general, these distinct principles for counting actions, noting that an account of how reason can give unity to action would require much more detail. Some of my traditionally formed Thomist friends are wary of “possible worlds” talk. I submit that they shouldn’t be. Although possible worlds ontologies have been proposed that are metaphysically objectionable, contemporary philosophers most often employ possible worlds language in a metaphysically innocent or neutral way, simply as a shorthand for considering various modal claims about the way things might be or might have been. In this common usage, a possible “world” is simply a possible “scenario.” One could argue about whether using possible worlds language

Notes

203

in this way improves economy of expression. However one answers that question, it remains the case that such usage has become conventional within contemporary philosophy, and my point here is simply that one can use the language without baggage to make or consider modal claims that any philosopher will want to make or consider. 27 Perhaps, if one has a Platonic view of relations, then one might think that relations, qua universals, are in some sense prior to their relata. But, a universal relation would clearly not be logically sufficient for A (the universal exists in all worlds, but, according to EM, God might not have caused A; so, given DUC’s claim that A can’t exist unless caused by God, A exists in only some worlds). Only a particular relation, or particular relation instance, could (arguably) be logically sufficient for A. But a particular relation, or relation instance, is not prior to A. So, whether we think of relations as universals or particulars, the causal relation between God and A will not constitute an item both prior to and logically sufficient for A. 28 We have already been introduced to the concept of a “basic action” in Section 1.1. See again Jennifer Hornsby and Naomi Goulder, “Action.” See also, E.J. Lowe, who defines a basic action as “an action in which the agent causes an effect of a certain kind, but not ‘by’ any means whatever.” E.J. Lowe, A Survey of Metaphysics (Oxford: Oxford University Press, 2002), 199–201. 29 Lowe, Survey of Metaphysics, 198–200, uses the example of an act that consists simply in an agent’s directly moving his hand. Timothy O’Connor maintains that an agent’s activity “par excellence” consists in his directly causing an intention to do something. See Persons and Causes, 72, n. 11. 30 For example, if y is a metaphysical part of x from which x derives (as in the relation of matter and form to a material substance on a hylemorphic metaphysics), then y will be causally and explanatorily prior to x. However, if, say, a reductive materialist holds that a mental state simply consists in a certain brain state, the mental state will presuppose the brain state, in the sense that it cannot be prior to that brain state, but neither will the brain state be prior to the mental state—rather, the claim would be that the mental state just is the brain state. 31 For a characterization of the agent’s action along these lines, see O’Connor Persons and Causes, 72, n. 11; see also, O’Connor, “Simplicity and Creation,” 407–09. 32 Notice that we can run the same argument for views that take an agent’s basic action to consist in the direct causing of a bodily movement. When, for example, Lowe (Survey of Metaphysics, 210) says that “A’s waving his hand was an action of A’s—it was A’s causing a certain kind of movement in his hand,” it is clear that the act in question consists either in the causal-dependence relation holding between A and the movement or in the movement together with that relation. Either way, the act cannot be explanatorily prior to the movement, since that in which the act consists (either the relation or the relation and the movement together) cannot be explanatorily prior to the movement. 33 Notice that the constant co-occurrence of A and God’s act of causing A is fully explained as a necessary consequence of the combination of DUC and EM. It follows from DUC and EM that it is not possible to have A without God’s act of causing A nor God’s act of causing A without A. There is no need to seek an explanation of their constant co-occurrence in anything further. 34 In earlier work on these issues, I sometimes speak of A as an essential constituent of God’s act of causing A. Even there, as the context should make clear, I do not mean to imply any priority of A to God’s act of causing A. But I now prefer simply to speak

204

Notes

of God’s act as essentially consisting in A and its dependence relation, for I think this latter way of speaking is less likely to mislead the reader into thinking that A is a metaphysically prior part from which God’s act derives. 35 Notice that many of the passages I quote in Section 1.2, as representing the prevailing opinion that God’s causing our actions is inconsistent with their being free in the libertarian sense, objected that God’s causing our actions would preclude our ability to do otherwise. 36 Timothy O’Connor emphasizes that this is the sort of ability to do otherwise endorsed by libertarians in “Free Will,” section 3.2. See also, Kane, A Contemporary Introduction, 38. As noted before, the terms “antecedent” and “prior” in my discussion mean antecedent and prior in the order of dependence, not just in the order of time. Indeed, arguably, it is the former order that is the important one. 37 I take “doing other than A” to be consistent with performing some act other than A or simply refraining from A. Here and in what follows, all doing and doing otherwise is voluntary, intentional, and rational, that is, motivated by reasons. 38 To keep the objection from being obviously implausible, we assume here all necessary conditions for S’s performance of A except S’s performance of A itself. 39 “Exists too late” does not mean primarily (or even necessarily) too late temporally, but rather that such a factor comes too late in the order of explanation, causality, or influence to make S perform A, or to place a limit or restriction on S’s power, such that S could not have done otherwise. 40 Here I use “presupposes” according to the meaning given in the previous section. 41 The example borrows an Aristotelian framework on which risibility is, for human beings, a necessary accident, one which a human being—and, I assume, a jolly old elf—could not fail to have. 42 Kane, Significance, 34–35. 43 Ibid., 35. 44 Though on 78, Kane appears to hold that satisfying UR is not just necessary, but also sufficient, for an act’s being up to its agent in the sense required for free will. 45 Though himself a libertarian, Kane states (Significance, 36) that the “ability to do otherwise” component of condition R is open to interpretation along compatibilist as well as incompatibilist lines. In showing that God’s causing our actions is consistent with our having the ability to do otherwise all antecedent conditions remaining the same, I have therefore shown that God’s causing our actions is consistent with a stronger sense of “ability to do otherwise” than that required by R. 46 Kane, Significance, 73. 47 Ibid., 34. 48 Ibid., 73. 49 Note that an act performed voluntarily, intentionally, and rationally where the agent had the ability, all antecedent conditions remaining the same, voluntarily, intentionally, and rationally to have done otherwise would appear to count as what Kane calls a “self-forming action,” a kind of action for which, on Kane’s view, we are ultimately responsible and from which ultimate responsibility for non-self-forming actions can derive. See A Contemporary Introduction, 120–31. 50 Denys Turner, Thomas Aquinas: A Portrait (New Haven: Yale University Press, 2013), 157–58. 51 Brian Davies, Thomas Aquinas on God and Evil (Oxford: Oxford University Press, 2011), 75. Emphasis added. For other contemporary representatives of this tradition, see, in alphabetical order, Boyle, Grisez, and Tollefsen, Free Choice, 97–103; David

Notes

205

B. Burrell, Freedom and Creation in Three Traditions (Notre Dame: University of Notre Dame Press, 1993), 111–18; Michael J. Dodds, Unlocking Divine Action: Contemporary Science & Thomas Aquinas (Washington, DC: The Catholic University of America Press, 2012), esp. 205–28; Harm J.M.J. Goris, Free Creatures of an Eternal God: Thomas Aquinas on God’s Infallible Foreknowledge and Irresistible Will (Leuven: Peeters, 1996), 276–304; (Goris’s title gave me the idea for the title of this chapter); Reinhard Hütter, Dust Bound for Heaven: Explorations in the Theology of Thomas Aquinas (Grand Rapids: Eerdmans, 2012), 249–82; Koons, “Dual Agency,” 397–410; Bernard J.F. Lonergan, Grace and Freedom: Operative Grace in the Thought of St. Thomas Aquinas (New York: Herder and Herder, 1971), esp. 103–09; R.J. Matava, Divine Causality and Human Free Choice: Domingo Báñez, Physical Premotion and the Controversy de Auxiliis Revisited (Leiden: Brill, 2016); Herbert McCabe, God Matters, 10–24; McCann, Creation and the Sovereignty of God, esp. 103–11; Miller, A Most Unlikely God, esp. 122–41; James F. Ross, “Creation II,” in The Existence and Nature of God, ed. Alfred J. Freddoso (Notre Dame: University of Notre Dame Press, 1983), esp. 128–35; Brian J. Shanley, “Divine Causation and Human Freedom in Aquinas,” American Catholic Philosophical Quarterly 72, no. 1 (1998): 99–122; and Kathryn Tanner, God and Creation in Christian Theology: Tyranny or Empowerment? (Minneapolis: Fortress Press, 2005), esp. 81–119 and 144–52. Those who make moves closest to ones made in this chapter are Boyle et al., Koons, and McCann, and, more recently, Matava. However, Peter Furlong (see his The Challenges of Divine Determinism: A Philosophical Analysis, forthcoming from Cambridge University Press) offers reasons for thinking that some of those cited above may end up holding deterministic views in contrast to my own. Furthermore, it would seem that the Báñezian Thomistic tradition, represented by Domingo Báñez, Reginald GarrigouLagrange, and more recently by Steven A. Long, David S. Oderberg, Thomas M. Osborne, Jr., and Joseph G. Trabbic, is not consistent with Dual Sources or with libertarian freedom more generally (though it still has many things in common with Dual Sources). For God’s physical premotion, by which God moves free creatures to act according to this tradition, would seem to constitute a factor both prior to and logically sufficient for creaturely actions, thus rendering them determined. Of course, proponents of this tradition may well be content with a compatibilist understanding of creaturely freedom in relation to God, and I leave open the possibility that I have misunderstood physical premotion. I make no claim about whether the Báñezian tradition adequately accommodates the divine transcendence. For recent proponents of this tradition, see Steven A. Long, “St. Thomas Aquinas, Divine Causality, and the Mystery of Predestination,” in Thomism and Predestination, ed. Steven A. Long, Roger W. Nutt, and Thomas Joseph White (Ave Maria, FL: Sapientia Press, 2016), 51–76; David S. Oderberg, “Divine Premotion,” International Journal for Philosophy of Religion 79, no. 3 (2016): 207–22; Thomas M. Osborne, Jr., “Divine Providence: Thomist Premotion and Contemporary Philosophy of Religion,” Nova et Vetera 4 (2006): 607–32; and Joseph G. Trabbic, “Praemotio Physica and Divine Transcendence,” also in Long, Nutt, and White, eds., Thomism and Predestination, 152–65. Oderberg offers some interesting discussion regarding how the thesis of physical promotion relates to libertarianism and compatibilism. 52 From Aquinas’s Commentary on Aristotle’s De Interpretatione, 1.9, in Thomas Aquinas: Selected Philosophical Writings. Selected and trans. Timothy McDermott (Oxford: Oxford University Press, 1993), 281. Emphasis added. 53 Ibid, 282–83.

206

Notes

54 And the conception that helps make sense of Aquinas’s claims elsewhere that God can cause acts of our will without rendering them determined or necessitated, or without precluding there being initiated by us, as well as by God. See, for instance, ST 1–2.10.4, quoted in section 1.3, and ST 1.83.1 ad 3. 55 There are ways of taking Aquinas’s claim that God’s will cannot fail (cannot be ineffectual) that would not be consistent with Dual Sources. It is certainly consistent with Dual Sources to say that God’s will, which is identical to God, causes any creaturely effect E. But it is not consistent to say that God’s will is a logically sufficient cause of E. That would result in E’s being determined or necessitated. It is consistent with Dual Sources to say that God’s willing E (where that is identical to God’s causing E) is logically sufficient for E. But it is not consistent to say that God’s willing E also causes E or is that in virtue of which God causes E. For that, likewise, would necessitate E. Similar points will be raised again in Chapter 8.

Chapter 5 1 2 3

4

5 6

7 8

With respect to (iii), I’m not sure what such a feature would be, but were an account of divine agency to include such a feature, it would count as an intrinsic model. For references, see Section 4.3. That God’s choice to cause E could not be impeded or ineffectual seems very plausible. That’s not to say that God couldn’t make a choice to cause E given certain conditions, conditions that might not be met. But, in a case where the conditions weren’t met, God’s choice would not be ineffectual, since he did not choose absolutely to cause E, but only to do so given the unmet conditions. For Aquinas on God’s freedom to do otherwise, see ST 1.19.3–4; ST 1.25.5 corpus and ad 1&3; ST 1.25.6 corpus and ad 6; SCG 2.23; and DP 3.15. Notice that the objection here has nothing to do with time or change. It is not the objection that God’s doing otherwise would require that God undergo change, in violation of God’s immutability and eternity. That is not a genuine problem, since, even though it is not possible for an unchanging God from all eternity to do not-A if from all eternity God is doing A, it is possible that, from all eternity, this very same God might have done not-A, instead. Rather the problem is that if God is identical to the act of doing A, then it is not possible for this God to exist without the act of doing A and hence not possible for this same God to have done not-A. This objection would seem to hold, at least if we assume “object essentialism,” to be discussed below. For discussion, see my and Mark K. Spencer’s “Activity, Identity, and God: A Tension in Aquinas and His Interpreters,” Studia Neoaristotelica: A Journal of Analytic Scholasticism 12, no. 2 (2015): 5–61. It is interesting to compare SIM and EM on this point. On EM, God’s activity of causing A is logically sufficient for A, but not prior to A, since God’s act partially consists in A. On SIM, by contrast, God’s act is prior to A, since this act is identical to God, who is causally prior to A. But, unlike on EM, God’s act is not logically sufficient for A, since it doesn’t have A as its object essentially and thus exists in some worlds in which A does not. EM clearly embraces object essentialism, since, as we know from Section 4.3, it takes God’s act of causing E to partially and essentially consist in E. SCG 2.11.

Notes

207

9 SCG 2.12.1–2. 10 See, for example, the passage from William Lane Craig, quoted and discussed in Section 5.4. 11 See Physics Bk. 3, 202a and Metaphysics Bk. 11, 1066a. 12 For Aquinas, see ST 1–2.110.2; ST 1.18.3 ad 1; Commentary on Aristotle’s Metaphysics 11.9.2309–2313. For Suarez, see DM 48.4.10–15 and DM 49.1.8. As Brian Davies explains, Aquinas presupposes the Aristotelian principle in his claim that God can act and cause without undergoing change. See Brian Davies, The Thought of Thomas Aquinas (Oxford: Oxford University Press, 1992), 112–13. The principle also seems in the background of passages such as DP 7.8, where Aquinas insists, with a view to preserving divine simplicity, that relation, action, and causality need involve no intrinsic state of the subject that enters into composition with the subject, or in virtue of whose gaining or losing the subject undergoes change when it begins or ceases to be related, to act, or to cause. 13 Freddoso, “Pitfalls and Prospects,” 139. 14 Ibid., 138. Freddoso also discusses the principle in “Introduction,” xlvi–xlvii. 15 Aquinas, DP 3.3. See also, SCG 2.18.2: “For creation is not a change, but the very dependency of the created act of being upon the principle from which it is produced. And thus, creation is a kind of relation; so that nothing prevents its being in the creature as its subject.” 16 It must be admitted that despite passages from Aquinas like the ones quoted and referenced above, there are a number of passages in which he appears to identify God’s acts that take creatures as objects simply with God or the divine substance, in a way that seems in tension both with the passages referenced here and with EM. This tension in Aquinas is reflected in conflicting interpretations among Thomists and interpreters of Aquinas. For detailed treatment of these interpretive difficulties, along with various possibilities for resolving them, see my (with Mark K. Spencer) “Activity, Identity, and God: A Tension in Aquinas and His Interpreters.” It is out of respect for these interpretive difficulties that I stop short of calling EM a “Thomistic” account of divine agency, even though one can point to passages in Aquinas that seem to support it. Briefly, one strategy for reconciling EM (or a close variant) with passages in which Aquinas identifies all of God’s activity with the divine substance picks up on Aquinas’s distinction at DP 3.3 between “creation taken actively” and “creation taken passively.” The former Aquinas identifies with the divine substance, while the latter he identifies with a real relation of the creature to God. But, presumably, Aquinas does not mean these to be two separate acts of creation. More likely, the taken actively taken passively distinction is just a way of talking about what is involved in a single act of creation on the side of the agent (God) and on the side of the patient (the creature). But, then, creation taken as a whole would not be just the divine substance but would include the creature’s relation of dependence on the divine substance. On this approach, when Aquinas identifies God’s acts that take creatures as objects with the divine substance, he is not referring to those acts taken as a whole but only to those acts taken actively, that is, to what they involve on the side of God, which is nothing in addition to God or the divine substance. Notice, moreover, that this understanding of what creation as a whole consists in for Aquinas would not constitute an intrinsic model of divine agency, for creation as a whole would not be intrinsic to God, even if creation taken actively just is God. Furthermore, this understanding of creation as a whole would appear to presuppose object essentialism, since were the creatures and their relations of dependence on God different, the act of creation taken as a whole would be different.

208

Notes

17 The conflict is obvious if God’s act of bringing about E, or choice to bring about E, is identical to God, or essential to God. But it will also be the case if there is some other intrinsic item that is identical to God, or essential to him, which would not exist were God not causing E. For, then, God would not exist were he not causing E, which clearly conflicts with the belief that God was free to do other than cause E. 18 The same is true of anything intrinsic to, but distinct from (not identical to), a thing. 19 In what follows I will speak in terms of “substance” rather than the more generic “subject,” but the gist of the argument is not affected by this choice. My discussion will be limited to objects whose composition, scholastics would say, includes just one substance, objects such as Socrates or Rover. If we assume that there are also objects whose composition includes more than one substance, call them aggregates (e.g., my computer), then one may ask whether such objects even have intrinsic accidents; if they do, the question of how those properties relate to such objects, and, in particular, whether they are included within such objects, becomes more complicated. Since, if God has intrinsic accidents, he is presumably more like Socrates (containing only one substance) than my computer (containing many), my purposes allow me to leave out discussion of the more complicated problem of the intrinsic accidents of aggregates. 20 Cf. Aquinas at ST 1.3.7, who in denying that God has component parts, gives as a reason that “every composite is posterior to its component parts, and is dependent on them; but God is the first being, as shown above.” 21 I have argued that where “substance” is opposed to accident as a subject in which accidents inhere, which itself excludes any accidents (call this substance-1), it is a mistake to identify Socrates, or anything else to which intrinsic accidents belong, as merely a substance. But “substance” can also be used to categorize objects in a way that does not leave out whatever accidents the objects have at a time. For instance, “substance” can refer to objects that persist through change, that exist of themselves and not simply in another, and that have irreducible powers. Socrates might be all of these things and hence a “substance” in this latter sense (call this substance-2), even though Socrates includes whatever accidents he has at a time (and thus is not to be identified with a substance-1). Both of these uses of substance are pervasive within the tradition. It is not surprising that the same word would be used with both significations, since a substance-2 will enjoy most of the characteristics that make it a substance-2 (e.g., the ability to maintain its identity across change) in virtue of its substance-1. I endorse both uses of “substance,” provided that we don’t allow them to confuse us into thinking that an object like Socrates is a substance-1, excluding whatever accidents he has at a time. 22 The metaphysics of persistence and change, substance and accident, and composition and identity that I have employed in the foregoing is certainly suggested by passages from Aquinas, though I don’t claim to have done the work here to establish that this is his position. But consider, for example, these lines from ST 1.9.1, where Aquinas gives the following reason for thinking God to be immutable: “Because everything which is moved, remains as it was in part, and passes away in part; as what is moved from whiteness to blackness, remains the same as to substance; thus in everything which is moved, there is some kind of composition to be found. But it has been shown above (Q. 3, A. 7) that in God there is no composition, for He is altogether simple.” The natural way to read these lines is as saying that a thing that undergoes change from being white to being black is composed, first of its substance and whiteness, and then of its substance and blackness. The changing thing, then, is not

Notes

23 24

25

26

27

209

just a substance (as subject of accidents) but something composed of substance and accident. See also, ST 1.3.3, where Aquinas indicates that accidents are included in the object which is a man: “Now individual matter, with all the individualizing accidents, is not included in the definition of the species. For this particular flesh, these bones, this blackness or whiteness, etc., are not included in the definition of a man. Therefore, this flesh, these bones, and the accidental qualities distinguishing this particular matter, are not included in humanity; and yet they are included in the thing which is a man.” For some twentieth-century scholastics whose metaphysics of these matters is (or appears to be) the same as what I have presented, see George P. Klubertanz, Introduction to the Philosophy of Being (New York: Appleton-CenturyCrofts, Inc., 1955), 64–80, 95, and 101; Leo Sweeney, Authentic Metaphysics in an Age of Unreality, second edition (New York: Peter Lang, 1993), 41–69; and Clarke, The One and the Many, 126–33, 150–58. See Aristotle’s Physics, Bk. 8, chs. 4 and 5. Aquinas comments favorably on this teaching in his Commentary on Aristotle’s Physics, Bk. VIII, lectures 10–11. That Aquinas endorses this teaching is made clear at SCG 1.13.21–SCG 1.13.23. On certain versions of this view, relations are understood to be universals jointly exemplified by the items related. I will assume a version of this account on which a relation that exists between two relata is not a universal, but rather a particular, and thus not numerically the same as any relation existing between any other pair of relata. It is worth noting that the medium view of relations is, with perhaps just one or a few obscure exceptions, a departure from the way the medievals and scholastics typically thought of relations. See, for example, Jeffrey Brower’s discussion of “the rejection of polyadic properties” in his “Medieval Theories of Relations,” in The Stanford Encyclopedia of Philosophy (Winter 2015 Edition), ed. Edward N. Zalta http://plato.stanford.edu/archives/win2015/entries/relations-medieval/. For a discussion of this approach, including a discussion of how proponents of this view dealt with the problem of merely Cambridge or relational change, see Brower, “Medieval Theories of Relations,” section 4.2. For philosophers who have interpreted Aquinas as holding a view along these lines, see Joseph Owen, An Elementary Christian Metaphysics (Houston: Center for Thomistic Studies, 1985), 179–90; and Leo Sweeney, Authentic Metaphysics in an Age of Unreality, 201–32. For the reductive view, see Brower, “Medieval Theories of Relations,” section 4.1. How to interpret Aquinas on relations is a controversial matter. While we’ve seen that some interpret him as holding the non-reductive view, for helpful presentations of the reductive view, presented as an interpretation of Aquinas, see T.M. Ward, “Relations without Forms: Some Consequences of Aquinas’s Metaphysics of Relations,” Vivarium 48, no. 3 (2010), 279–301; and Mark G. Henninger, “Aquinas on the Ontological Status of Relations,” Journal of the History of Philosophy 25 (1987), 491–514. Describing Aquinas’s view, Ward notes (282) that “in addition to absolute accidents, substantial forms and supposits can also serve as the foundations of real relations.” As an example of substantial forms serving as foundations, Ward says (282) that it is the substantial forms of Socrates and Plato that serve as the foundation for their relations to one another of sameness of species (282). As examples of whole supposits (or subjects) serving as foundations, Ward says (296) that if b is the son of a, the foundation of the relation of sonship is the whole supposit b and that the foundation for the creature’s dependence relation to God is the whole creaturely supposit.

210

Notes

28 For example, Aquinas denies that an object known is really related to the knower simply in virtue of being known. If Cecilia knows Elizabeth, it will be true that “Elizabeth is known by Cecilia.” That truth will not, however, involve any entity in Elizabeth that is Elizabeth’s relation to Cecilia of being known by her. Rather, the reality that makes the statement “Elizabeth is known by Cecilia” true will be simply Elizabeth, Cecilia, and a certain mental or cognitive state of Cecilia that, under the right conditions, constitutes, or has the ratio of, a relation to Elizabeth of knowing her. 29 Where that intrinsic with respect to a subject that relates the subject to another is the subject itself, I think it is more natural to speak of the subject (in this case E) relating itself to another (in this case God) than to speak of E as the foundation for a causal-dependence relation to God and thus as constituting its own relation to God. The former manner of speaking is consistent with denying that there is any entity that is the relation of E to God, implying that such an entity isn’t necessary where E relates itself to God. The latter manner of speaking affirms that there is an entity that is the relation of E to God, but the relation turns out to be E itself. Thus, the latter manner of speaking flouts our normal expectation that where there is an entity that is x’s relation to y, that entity is distinct from, not merely identical to or constituted by, x. In short, speaking of something intrinsic to a subject as constituting that subject’s relation to another is best reserved for cases in which the intrinsic item in question is not the subject itself. Where the item is the subject itself, it is more felicitous to speak of the subject’s relating itself to another. On either way of speaking, however, the substance of the reductive view is the same for cases in which a subject relates itself to another: There is no entity existing separately from the subject that is the subject’s relation to the other. 30 Assuming the scholastic teaching that God is not really related to creatures, although on the non-reductive view God’s causing E will involve a relational entity in E—E’s relation of causal dependence on God—it will not involve any correlative relational entity in God. The ontological basis for God’s relation to E as E’s cause is E’s real relation of causal dependence on God. God’s relation to E as cause is, thus, understood to be a “rational” not a “real” relation, not because God isn’t truly related to E as cause but because the ontological basis for that truth does not include an intrinsic entity in God that is God’s relation to E. 31 Suarez adopts a similar strategy in response to the objection that God’s act of creation would need to be created by him, thus launching an infinite regress. Says Suarez (DM 20.4.31, 89): “After all, the same regress could be inferred in the case of any action whatsoever. For every action is in its own way an effect of the agent from which it flows; yet it is not effected, nor does it flow, by means of another action, since it flows not as an effect, but as the very outflowing and dependence itself. So, then, in a broad sense an act of creation can be called a creature, since it itself is also effected by God ex nihilo. Yet it is not a creature absolutely speaking, in the way that the entity which is made and serves as the terminus of God’s production is; rather, it is a creature in the way that a dependence or production itself is—and so no other act of creation is required for it.” 32 Of course, a primary reason to deny that a thing’s having a relation to something introduces a new relation of “having” between the thing and the first relation is that such a position leads to an infinite regress. To avoid the regress, we don’t deny the relational truth that a thing has its relations, but we do deny that having a relation introduces a new relation distinct from the relation had. Cf. Aquinas, DP 3.3 ad 2,

Notes

33

34

35

36 37 38

211

who when speaking of the creation relation says: “Nor is it necessary to proceed to infinity, since the relation of creation is not referred to God by another real relation but by itself: because no relation is related by another relation, as Avicenna says (Metaph. iii, 10).” This is certainly O’Connor’s view; see Persons and Causes, 72, n. 11, where O’Connor says that an agent’s activity par excellence consists in a causal relation to a determinate state of intention. In cases of observable bodily action, like waving a hand, O’Connor holds that the agent’s act consists in the relation to the intention, plus the bodily states that flow from that intention. We could say, then, that the agent’s basic action consists in a causal relation to the intention. As noted when presenting EM in Section 4.3, I think it makes most sense to say that the effect is included in the agent-causal act, for if an agent-causal act were merely a causal relation, then they would all seem to be acts of the same type. This point would not seem critical to the present argument. To say that an agent causes its action seems very natural, and a number of authors have affirmed it, though for various reasons, and despite holding significantly different accounts. To cite just a few examples, see Taylor, Action and Purpose, 112; Hasker, The Emergent Self, 104–05; and Randolph Clarke, Libertarian Accounts of Free Will (Oxford: Oxford University Press, 2003), 148–49. (i) holds that in directly causing or bringing about an effect, an agent-cause also causes or brings about the causal relation that holds between it and the effect. But does the agent-cause directly or basically agent-cause that causal relation? Just above we characterized a basic agent-causal act as consisting in a causal relation R to its effect E. It follows that if the agent-cause agent-causes R in a basic agent-causal act, that will introduce a new relation, R*, between the agent and R. And, if the agentcause agent-causes R* in a basic agent-causal act, that will introduce yet another relation, R**, and so on to infinity. But one of the reasons for accepting (i) is to avoid such infinite regresses. We can do so by denying that the agent-cause’s causing of R is a basic agent-causal act. Rather, the agent-cause’s causing of R is part of the basic agent-causal act of agent-causing E. In the basic agent-causal act of causing E, the agent-cause also brings about R, but this bringing about of R is not a new agentcausal act, distinct from the act of bringing about E. The qualification “that take relata” is made in order not to preclude accounts of the Trinity on which the Divine Persons are understood to be relations that do not have relata. To be clear, this claim is consistent with R’s being brought about by the same agent that performed the act of which R is a part. Freddoso (“Introduction,” xcviii; and “Pitfalls and Prospects,” 151–56) maintains that according to classical concurrentists such as Aquinas and Suarez, though God and the creaturely agent bring about their common effect by different powers, they do not do so in two separate acts, but rather share a single act in bringing about the effect. Notice that the position Freddoso describes seems to follow if we assume the reductive view of relations and the Aristotelian principle that the action of the agent is in the patient. For, if (as the principle claims) the act of an agent just is its effect qua dependent on that agent, and if (as the reductive view claims) the dependence relation of the effect on the agent is not an entity distinct from the effect, then it follows that the entity that constitutes a creature’s act of causing an effect E and the entity that constitutes God’s act of causing that same effect E is one and the same entity. In other words, the creature’s act of causing E and God’s act of causing E will

212

39 40 41

42

43

44

45

Notes refer to the same thing. On the other hand, given the medium and non-reductive views, God’s act and the creature’s act will not refer to the same thing. For, God’s act will refer to E plus E’s distinct dependence relation to God, and the creature’s act will refer to E plus E’s distinct dependence relation to the creature. Since I am allowing EM to be construed according to any of the three views, I understand EM to be open to either way of answering the question whether God and a creaturely agent share a single act in bringing about their common effect. As is the case throughout this book, “prior” in the context of this discussion means prior in the order of causality, explanation, dependence, or influence, not prior in the order of time. William Lane Craig, “Creation, Providence, and Miracles,” in Philosophy of Religion: A Guide to the Subject, ed. Brian Davies (Washington, DC: Georgetown University Press, 1998), 142–43. Notice that whether x would be intrinsically the same after causing y, or even simultaneously with causing y, is irrelevant to the question of whether x had what it takes to explain, or causally account for, the existence of y. For what comes after or simultaneous in the order of explanation comes too late to play any explanatory role in accounting for the causing or the effect caused. Of course, to deny that God has intrinsic accidents is to deny that God would be intrinsically different before, simultaneously, or after causing an effect. But God’s state simultaneous to or after the effect is produced has no significance for whether it makes sense to say that God caused or accounted for the effect. If one thinks that God’s having produced a contingent effect requires that God simultaneously or subsequently acquire an intrinsic accidental property of having produced that effect, then one will have in mind a different and, it seems to me, far less interesting objection. Related questions about how the denial of intrinsic accidents in God squares with God’s knowledge of contingent truths will be taken up in Chapter 8. For others who have made points similar to the ones that follow, see Brower, “Divine Simplicity and Divine Aseity,” O’Connor, “Simplicity and Creation,” Pruss, “Two Problems of Divine Simplicity,” and my “Must a Cause Be Really Related to Its Effect?” Thus, even libertarians who argue that Frankfurt-style scenarios show that ability to do otherwise is not necessary for free will still tend to think that some, perhaps most, free actions are ones for which the agent has the ability to do otherwise. See, for instance, Eleonore Stump, “Libertarian Freedom and the Principle of Alternative Possibilities,” in Faith, Freedom, and Rationality: Philosophy of Religion Today, ed. Jeff Jordan and Daniel Howard-Snyder (Lanham, MD: Rowman & Littlefield, 1996), 88. And even libertarians who think that an action can be ultimately up to its agent even if it is determined by proximate causes still tend to think that ultimate responsibility must be rooted in self-forming actions for which the agent had the ability to do otherwise, all antecedent conditions remaining the same. See, for example, Robert Kane, Contemporary Introduction, 130–31. For a helpful introduction to these accounts, see Robert Kane, “Introduction: The Contours of Contemporary Free Will Debates,” in The Oxford Handbook of Free Will, 22–26; and, in the same volume, Randolph Clarke “Libertarian Views: A Critical Survey of Noncausal and Event-Causal Accounts of Free Agency,” 362–76. Anscombe, “Causality and Determination,” 136. For a concurring opinion, see Peter van Inwagen, An Essay on Free Will, 138–40.

Notes

213

Chapter 6 1

This despite the fact that, as we saw in Section 2.1, scripture sometimes speaks in a way that suggests that God is in some way causally behind sinful acts. For further discussion, see Section 7.6. 2 See, for example, Aquinas, ST 1.4.2. 3 The translation is my own, from the Latin found in Decrees of the Ecumenical Councils, Volume II: Trent to Vatican II, ed. Norman P. Tanner (London: Sheed & Ward, 1990), 679. The Latin for the entire sixth canon reads: “Si quis dixerit, non esse in potestate hominis vias suas malas facere, sed mala opera ita ut bona Deum operari, non permissive solum, sed etiam proprie et per se, adeo ut sit proprium eius opus non minus proditio Iudae quam vocatio Pauli: a.s.” See also, the Westminster Confession (1646), chs. 3.1 and 5.4, and the thirteenth article of the Belgic Confession (1561). 4 Anselm, On the Fall of the Devil 20, 223. See also Anselm’s De concordia 1.7. 5 Aquinas, ST 1–2.79.2. Earlier in the same article, Aquinas had remarked, “The act of sin is both a being and act; and in both respects it is from God. Because every being, whatever the mode of its being, must be derived from the First Being, as Dionysius declares.” Aquinas makes this same point at De malo 3.2. 6 Aquinas, ST 1–2.79.2 ad 2. 7 Their approach was, in fact, common within the scholastic tradition and was also endorsed by Descartes, for instance, in his Fourth Meditation. 8 See De malo 1.3 ad 13, and De malo 2.2. 9 We have seen that Aquinas wants to deny that God causes sins. But some readers, sympathetic with the general thrust of Aquinas’s solution, may be happy to allow that God causes “sin,” where “sin” denotes only what I’m calling the act of sin and not also the privation in which the act’s sinfulness consists. Such readers will argue that there is no problem in God’s causing “sin,” so understood, provided that God not cause an act’s sinfulness or that in virtue of which it is a sin. While a reader who takes this approach will not find the conclusion of the argument set out above problematic, he will still make use of our solution to reconcile DUC with his denial that God causes that in virtue of which a sin is sinful. 10 If defensible, the privation solution not only enables us to block the inference from DUC to God’s causing sin. For those wishing to deny that God causes sin, it also provides a potential way of making sense of passages in scripture, noted in Sections 2.1 and 7.6, that describe God not merely as allowing or permitting sinful acts but as actively involved in their production. Given the privation solution, one might say that God is actively involved in the production of sinful acts, since he causes every act of sin; yet God does not cause sin, since he does not cause the defect in which an act’s sinfulness consists. 11 The passage is from Aquinas’s De ente et essentia, 1, 91–92. 12 Accordingly, someone uncomfortable with the language “X exists” or “There is an X,” where X does not name an entity, can understand “X exists” or “There is an X” as paraphrases of “There does not exist some Y.” For example, “There is a lack of conformity to the moral standard” could be understood as a paraphrase of “There is not a relation of conformity to the moral standard.” Of course, to characterize the lack of conformity as not just a lack but also a privation is to say that the relation of conformity to the moral standard ought to exist in the act so deprived (or between the act and the standard).

214

Notes

13 See Patrick Lee, “The Goodness of Creation, Evil, and Christian Teaching,” The Thomist 64, no. 2 (2000): 239–69; and Patrick Lee, “Evil as Such Is a Privation: A Reply to John Crosby,” American Catholic Philosophical Quarterly 81, no. 3 (2007): 469–88. See also, Bill Anglin and Stewart Goetz, “Evil Is Privation,” International Journal for Philosophy of Religion 13, no. 1 (1982): 3–12; and David E. Alexander, Goodness, God, and Evil (New York: Continuum, 2012), esp. chs. 2–4. 14 See John F. Crosby, “Is All Evil Only Privation?” Proceedings of the American Catholic Philosophical Association 75 (2002): 197–209; John F. Crosby, “Doubts about the Privation Theory That Will Not Go Away: A Response to Patrick Lee,” American Catholic Philosophical Quarterly 81, no. 3 (2007): 489–505; Todd C. Calder, “Is the Privation Theory of Evil Dead?” American Philosophical Quarterly 44 (2007): 371–81; and G. Stanley Kane, “Evil and Privation,” International Journal for the Philosophy of Religion 11 (1980): 43–58. 15 Gottfried Wilhelm Leibniz, Confessio Philosophi: Papers concerning the Problem of Evil, 1671–1678. Ed. and translated by Robert C. Sleigh, Jr. (New Haven: Yale University Press, 2005), 113. For helpful discussion of Leibniz’s views on the privation account of evil, see Samuel Newlands, “Leibniz on Privations, Limitations, and the Metaphysics of Evil,” Journal of the History of Philosophy 52 (2014): 281–308. 16 Leibniz, Confessio Philosophi, 111. 17 Ibid., 113. 18 Newlands, “Leibniz on Privations, Limitations, and the Metaphysics of Evil,” 288. 19 For my purposes “sinful act” and “morally bad act” can be used interchangeably. By “active power” I mean a power that an agent has to perform some act or operation. Some might also admit “passive powers,” which are understood as an agent’s power to be acted on, without thereby becoming the subject of an act or operation itself. By the adjective “active” I mean to distinguish the kind of power that concerns us here from merely “passive powers” so understood. 20 One of the things that makes this assumption controversial, of course, is that it implies commitment to something like Aristotelian natural teleology, including the claim that we can speak sensibly of unconscious entities having natural “ends,” “goals,” or “aims,” the achievement of which is a good for those entities. While I am unable fully to defend such foundational commitments here, it is worth noting the renewed interest in such ideas by analytic philosophers working on the notion of powers. George Molnar, for example, writes that “powers, or dispositions, are properties for some behavior.… These properties have an object towards which they are oriented or directed” (60). Thus, “something very much like intentionality is a pervasive and ineliminable feature of the physical world” (61). “Physical powers, such as solubility or electromagnetic charge.… have that direction toward something outside themselves that is typical of psychological attributes. The intentional object of a physical power is its proper manifestation. Of the many ways of characterizing a power, the only one that reveals the nature (identity) of the power is the characterization in terms of its manifestation” (63). See George Molnar, Powers: A Study in Metaphysics (Oxford: Oxford University Press, 2003). 21 This point is true even if there are positive properties, for instance, of the eyes, which help explain why my seeing falls short of the standard. 22 Aquinas makes the distinction between “human acts,” of which humans are master by reason and will, and “acts of a human,” which do not proceed from reason and will, at ST 1–2.1.1.

Notes

215

23 See P.T. Geach, “Good and Evil,” Analysis 17 (1956): 33–42. See also, Bernard Williams, Introduction to Ethics (New York: Harper and Row, 1972), 40–50. For an excellent recent defense of this view, see Alexander, Goodness, God, and Evil, esp. chs. 2–4. 24 For helpful defenses of the privation account of evil, in general, see the literature cited in note 13. The considerations of this and the following paragraph are developed in more detail by Lee and Alexander. 25 For instance, the exercise of active power is good insofar as it achieves, at least to some extent, that which the power is a power for. 26 To be clear, I am not using the privation account of evil in general as a premise in an argument for the privation account of moral evil. Rather, I am pointing to an advantage in saying that the badness in a sinful act consists in a lack of conformity to the moral standard as opposed to consisting in some positive property of “conflicting with the moral standard.” The advantage is that the former approach is consistent with a privation account of evil in general, an account that is attractive for the reasons stated. 27 As we will note in reply to the third objection in the next section, the privation account of moral evil is even consistent with calling properties in respect of which an act lacks conformity to the moral standard “morally bad.” It is only inconsistent with holding that the badness in a sin of action consists in these properties. 28 I consider two additional objections, which seem to me less threatening, in my “The Privation Account of Moral Evil: A Defense,” International Philosophical Quarterly 55, no. 3 (2015): 271–86. 29 Calder, “Is the Privation Theory of Evil Dead?” 372–73. 30 Ibid., 375. 31 For a discussion that highlights Aquinas’s recognition of the positive element in sinful acts, see Gregory M. Reichberg, “Beyond Privation: Moral Evil in Aquinas’s De Malo,” The Review of Metaphysics 55 (2002): 751–84. 32 Kane, “Evil and Privation,” 51–52. 33 Calder, “Is the Privation Theory of Evil Dead?” 374. The response by Anglin and Goetz to Kane can be found in their “Evil Is Privation,” cited above. 34 Crosby, “Is All Evil Only Privation?” 203. 35 Ibid., 204. 36 I do not have a particular view about what sort of thing “opposition to Abel’s life” is. What I am arguing is that if one wishes to call “morally bad” Cain’s act’s “opposition to Abel’s life,” where this is understood as a positive property or feature of Cain’s act, one can do so without contradicting the privation account of moral evil. 37 As is probably clear, I am assuming cases in which R, or the lack of R, or the truth aRb holds in virtue of a and b and their relevant properties, where R is not a causal relation holding between a and b, with one of the relata causing the other. Contrary to the sort of cases I have in mind here, I argued in Section 5.3 that it makes very good sense to say that an agent-cause brings about the causal relation between itself and its immediate effect in the act of causing that effect (even though the agent causes only the effect and not also itself). 38 Perhaps, it goes without saying that the applicable sense of “ought” varies among these examples. Indeed, the variety of examples illustrates that a number of different senses of “ought” can help ground causal explanations by nonperformance: moral oughts, prudential oughts, oughts used for the behavior that would be expected of a good or healthy member of its kind in the circumstances in question; oughts used

216

Notes

for the activity that a thing has been designed to perform, or that a thing has been adopted, as an instrument, to perform; etc. 39 Aquinas, ST 1–2.6.3. Aristotle also seems to recognize causation by nonperformance, albeit less explicitly. In Metaphysics, Bk. 5, ch. 2, Aristotle holds that that which when present is the cause of some particular effect is, when absent, the cause of the contrary effect, and he gives as an example a ship’s safety being caused when the pilot is present and its loss being caused when the pilot is absent. Presumably, if the presence of an agent explains some effect and the agent’s absence the opposite effect, it is only because when present the agent performs and when absent the agent does not perform. 40 This seems also (at least roughly) to be Aquinas’s account of how the sinner causes the privation in virtue of which his act is sinful. Thus, at ST 1.49.1 ad 3: “In voluntary things the defect of the action comes from the will actually deficient, inasmuch as it does not actually subject itself to its proper rule.” And ST 1–2.75.1: “The will lacking the direction of the rule of reason and of the Divine law, and intent on some mutable good, causes the act of sin directly, and the inordinateness of the act, indirectly, and beside the intention: for the lack of order in the act results from the lack of direction in the will.” See also, De malo 1.3; ST 1–2.75.1 ad 3; and SCG 3.10. Aquinas predominantly thinks in terms of the first way discussed above, in which the defect is imputable to the sinner in virtue of what the sinner doesn’t do, the sinner’s nonconsideration or nonuse of the moral rule. See my “Aquinas on How God Causes the Act of Sin without Causing Sin Itself,” The Thomist 73, no. 3 (2009): 455–96.

Chapter 7 1

2 3 4

5

6

Arguably, these conventional definitions need revision, reserving “moral evil” for lacks of conformity to the moral standard in things that ought so to conform, or for the things that lack conformity to the standard, along the lines described in the previous chapter. Other sorts of deprivation, as well as suffering, would then count as natural evil, even if they resulted from morally bad actions. Fortunately, nothing in what follows hinges on these details of definition. Richard Swinburne, Is There a God? (Oxford: Oxford University Press, 1996), 98. In the hands of a Molinist, FWD will have to be adapted somewhat from the description offered here. I will discuss Molinism in Section 7.3. An exception is Jason Turner, “Compatibilism and the Free Will Defense,” Faith and Philosophy 30, no. 2 (2013): 125–37. Turner argues that what he calls an “independent compatibilist” could adopt FWD by making it a necessary condition of an act’s being free that it not issue from a deterministic chain knowingly initiated by another agent (such as God), even though he allows that free acts can result from other sorts of deterministic chains. As they do throughout the book, terms like “antecedent,” “prior,” and “exists too late” signify not primarily (or even necessarily) temporal relations, but rather relations in the order of causal or explanatory dependence. For the detailed account of which the preceding paragraph is a brief summary, see Chapter 4, especially Sections 4.4 and 4.5. Here, to avoid additional complications and qualifications, I am assuming a strict libertarianism on which no act free in the libertarian sense can be determined.

Notes

217

But as discussed in Chapters 1 and 4, I, in fact, recognize the existence of broad libertarianism, in which some free acts are determined, such that the agent does not have the ability to do otherwise all antecedent conditions remaining the same. 7 To spell the point out: If the relations are possible, then, contrary to FWD Assumption, possibly, if in certain antecedent conditions I was free in the libertarian sense either to perform or refrain from some sinful act, and I freely (in the libertarian sense) performed it, then God could have done something such that, necessarily, had he done it, I would have freely (in the libertarian sense) refrained from the act in those conditions. 8 See Timpe, Free Will, 9. 9 That FWD Assumption (on which FWD depends) is false is a significant conclusion. One who acknowledges the conclusion might attempt to repair the perceived damage by substituting something else for the role played in FWD by libertarian freedom and moral responsibility. For example, one might replace FWD with an “Independence from God” defense. Such a defense would claim that it is a great good for human beings, not merely to have libertarian freedom, moral responsibility, and the ability freely to love God without necessitation or coercion—all things Dual Sources can provide—but to be able to perform acts independently of God—acts that are not caused by God, but are rather up to and under the control of the human being alone (even though being under the control of the human being alone is not necessary for genuine freedom and moral responsibility). While God can’t give us this sort of independence and at the same time ensure that we never sin, he is nevertheless justified in giving us the independence because it, and not merely genuine freedom and responsibility, is such a good thing to have. What should we say about this alternative to FWD? I think that for many, being able to act in independence from God (if judged a good at all) would be deemed a far less significant good than libertarian freedom and moral responsibility, once it is recognized that one can have freedom and ultimate responsibility and control over one’s actions without the independence from God. If that’s right, the “Independence from God” defense will be less attractive than the original FWD. That the “Independence from God” defense is incompatible with the traditional theistic doctrine of DUC goes without saying. 10 See, for example, Davies, The Reality of God and the Problem of Evil. See also, though not in a particularly scholastic vein, Charles Lewis, “Divine Goodness and Worship Worthiness,” International Journal for Philosophy of Religion 14, no. 3 (1983): 143–58. I think the views of these former and esteemed teachers of mine (Lewis’s may have since changed) ought to be given consideration, despite the fact that I assume the (at least currently) more common view here. 11 See Swinburne, Is There a God? 109–10, and Providence and the Problem of Evil, 160–71. 12 Swinburne, Is There a God? 109–10. 13 Cf. Aquinas, who at ST 1.48.2 ad 3, with an eye toward the sort of examples mentioned above, observes that “many good things would be taken away if God permitted no evil to exist” (see also, ST 1.22.2 ad 2 and SCG 3.71.6). Notice the claim here is not the (I believe) false one that moral goodness presupposes moral evil. It is only that certain sorts of goods (including certain exemplifications of moral goodness) presuppose moral evil. Edward Feser has put the point well: From the fact that God could have created a world where free creatures make morally significant choices, all of them good, “it doesn’t follow” that “God could have created a world with free will, no evil, and all the moral good that actually exists in the world. For

218

14

15

16

17

18

Notes there are still certain kinds of exercise of free will that presuppose the existence of people who choose evil.” Feser goes on to give examples of the sort I give above. See his Five Proofs of the Existence of God (San Francisco: Ignatius Press, 2017), 296. The NRSV’s “has imprisoned all in disobedience” replaces the RSV’s “has consigned all men to disobedience.” Notice that whether “all” is interpreted to mean literally all human beings or merely all classes of human beings (e.g., Jew and Gentile), the point remains that Paul seems to be teaching that God permits (or something stronger than permits) disobedience in order that he might be merciful. For discussion, see James D.G. Dunn, Word Biblical Commentary, Vol. 38b, Romans 9–16 (Dallas: Word Book Publishers, 1988), 693–97. For others who have proposed similar reasons for God’s permission of moral evil, see Paul Helm, The Providence of God (Downers Gove, IL, InterVarsity Press, 1993), 193–216, and Alvin Plantinga, “Supralapsarianism, or ‘O Felix Culpa’,” in Christian Faith and the Problem of Evil, ed. Peter van Inwagen (Grand Rapids, MI: Wm. B. Eerdmans Publishing Co., 2004), 1–25. Plantinga writes: “I believe that any world with incarnation and atonement is better than any world without it” (10), noting that “of course all worlds with incarnation and atonement contain evil. For atonement is among other things a matter of creatures’ being saved from the consequences of their sin.… a necessary condition of atonement is sin and evil” (12). Commenting on the atoning work of the Holy Trinity in the face of human depravity, Plantinga asks, “Could there be a display of love to rival this? More to the present purpose, could there be a good-making feature of a world to rival this” (7)? Plantinga (18) speculates, furthermore, that the relationship between God and a redeemed sinner has its own characteristic good or intimacy that would not be found in loving relationships between God and those who are not redeemed. And both Plantinga (18–19) and Marilyn McCord Adams propose that suffering (and one might think especially suffering at the hands of wrongdoers) affords an extraordinary kind of fellowship, identification, and solidarity with the atoning Christ and even enables the sufferer to display more fully the image of God in Christ. These, then, are also goods for the sake of which God might permit moral evil. See Marilyn McCord Adams, Horrendous Evils and the Goodness of God (Ithaca, NY: Cornell University Press, 1999), 166–68. The idea that evil makes possible the display of God’s goodness in creation has affinities with what Daniel M. Johnson calls “the divine glory defense.” For a helpful discussion of this defense, see Johnson’s “Calvinism and the Problem of Evil: A Map of the Territory,” in Calvinism and the Problem of Evil, ed. David E. Alexander and Daniel M. Johnson (Eugene, OR: Pickwick Publications, 2016), 43–48. To be clear, on the assumption that middle knowledge is possible (which I concede here for the sake of argument), I am not denying that it is logically possible that God’s middle knowledge is such that any free creature he could have created would have sinned. For this reason, I allow that the hypothesis can be used in a refutation of the logical argument from evil. What I am claiming is that the hypothesis seems so improbable that it defies belief. A response to the merely logical problem of evil need not be believable at all. That the values of worlds might be incommensurate in this way has been proposed by Thomas V. Morris and William Hasker. See Thomas V. Morris, “Perfection and Creation,” in Reasoned Faith, ed. Eleonore Stump (Ithaca: Cornell University Press, 1993), 236–37; and William Hasker, The Triumph of God over Evil: Theodicy for a World of Suffering (Downers Grove: InterVarsity Press, 2008), 79, 89–92.

Notes

219

19 Hasker sympathetically discusses this sort of argument against the claim that God must create the best possible world in The Triumph of God over Evil, 79, 85–92. 20 Although they do not show awareness of Dual Sources, Open Theists certainly claim an advantage over Molinism when it comes to addressing the problem of moral evil. See William Hasker, Providence, Evil and the Openness of God (New York: Routledge, 2004), especially 109–24; and The Triumph of God over Evil, especially 173–76 and 203–07. See also Alan R. Rhoda, “Gratuitous Evil and Divine Providence,” Religious Studies 46, no. 3 (2010): 281–302. 21 Neal Judisch reaches a similar conclusion in an argument with some affinities to the one that follows. See his “Meticulous Providence and Gratuitous Evil,” Oxford Studies in the Philosophy of Religion 4 (2012): 67–85. 22 Hasker, God, Time, and Knowledge, 192. 23 I call them “standard” because they are affirmed by Hasker, who is more than any other philosopher identified with the movement of Open Theism. 24 Hasker says (Providence, Evil, and Openness, 101) that, although Open Theists deny exhaustive divine foreknowledge, they hold that “God has a vast amount of knowledge about the probabilities that free choices will be made in one way rather than another.” 25 Despite the comparatively stronger picture of divine providence that might be gleaned from the passage quoted from Hasker above, when facing the question of God’s permission of sin, Hasker sometimes speaks in ways that suggest what I’m calling More Risky Open Theism. See, for example, The Triumph of God over Evil, 205. I don’t know whether More Risky Open Theism strictly contradicts the previous Hasker passage. But they do, certainly, give different impressions about the level of influence God can exercise over creation. 26 For a suggestion along these lines, see Swinburne, Providence and the Problem of Evil, 203–12. 27 Swinburne, Is There a God? 101. 28 If you doubt that a parent’s making his or her presence known would make misbehavior less likely, but not rule it out altogether, I challenge you to do some experimentation with your own kids! 29 One might argue that these points about humanity’s prelapsarian condition only show that God’s influencing things in the way I’ve suggested wouldn’t help much: after all, our first parents sinned despite these favorable conditions! Yet, that our first parents sinned in these conditions is, famously, not at all what one would have predicted. By contrast, it is highly predictable that we will sin in the conditions in which we currently find ourselves. The claim is not that divine influence of the sort I’ve suggested would guarantee no sin but only that it would almost certainly decrease its frequency and severity, giving rise to the question why God permits much sin that he could almost certainly have prevented. 30 The judgment whether it would be worth it may be affected, of course, by whether moral evils are compensated for, or defeated, by goods they make possible. Notice, however, that if the More Risky Open Theist were to say that God permits moral evils for the sake of goods they make possible, he would be giving the same justification that we’ve said needs to be given by the Dual Sourcer, Molinist, and Less Risky Open Theist. We will discuss further how compensating goods might play a role in More Risky Open Theism, below. 31 For an introduction, with lots of helpful references, see Daniel Howard-Snyder and Adam Green, “Hiddenness of God,” in The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), ed. Edward N. Zalta, http://plato.stanford/edu/archives/ win2016/entries/divine-hiddenness/.

220

Notes

32 Certainly, if natural regularity required something like the causal closure of the natural, such that it would be undermined by God’s bringing about effects in the natural world, then the divine influence suggested would admittedly undermine natural regularity. But that would be a much stronger conception of natural regularity than what’s needed to make the world intelligible, amenable to scientific study, and sufficiently predictable for the purposes of practical planning and deliberation. It would surely be a much stronger conception of natural regularity than what most theists accept. 33 For some discussion of the biblical conception with references to critics of Open Theism, see Section 2.1. See also, Hugh J. McCann and Daniel M. Johnson, “Divine Providence,” in The Stanford Encyclopedia of Philosophy (Spring 2017 Edition), esp. section 3, ed., Edward N. Zalta, http://plato.stanford.edu/archives/spr2017/entries/ providence-divine/. 34 See Hasker, in Clark Pinnock and others, The Openness of God, 152–54. 35 For worries that Open Theism makes God’s creation of free creatures an act too risky for an all-good Creator, see Flint, Divine Providence, 107; and McCann and Johnson, “Divine Providence,” section 3. 36 One is reminded of the quip that insanity is doing the same thing over and over again and expecting a different result. Peter van Inwagen has suggested that God’s love for humanity might lead him to attempt to rescue us from our depraved state of separation from him, rather than simply bringing the story of humanity to a close. See The Problem of Evil (Oxford: Oxford University Press, 2006), 87. The question is whether the God of More Risky Open Theism could have any confidence in his ability to rescue humanity. Either this God has been trying everything in his power to rescue humanity or he hasn’t. If he hasn’t, then his permission of the continued separation (with its attendant moral depravity) is unjustified absent compensating goods that moral evil makes possible. If he has, then with each passing generation he should be less confident in his prospects of achieving the hoped-for rescue. In such a situation, while it would make sense for God to attempt to rescue those human beings he has already brought into existence, it is not at all obvious that it would make sense for him to continue bringing them into existence, absent compensating goods. 37 This is certainly true for Dual Sources, but it is also true for Molinism, once we set aside what in Section 7.3 I call the “incredible hypothesis” that God’s middle knowledge so constricts God’s options that it was not possible for him to create a world with free creatures who never sin or who sin far less often and badly than the free creatures of our world. 38 The parenthetical remark is important. The God of More Risky Open Theism knows all too well by now (even if he didn’t know it at the beginning of creation) that the tendency of free creatures is to sin badly and often. Thus, what (a) represents is not the good of free creatures who might possibly use their freedom for bad, but rather the continuation of free creatures who almost certainly will use their freedom for bad. 39 See Hasker, Providence, Evil and the Openness of God, 115–16. 40 Ibid., 133. 41 Even so, on this account, it is hard to find the kind of distance between God and the occurrence of sin that the Open Theist would like to claim. For it remains the case, on this view, that God ultimately approves of a world mired in sin—a world that he could always bring to a close if he didn’t approve of it—largely because it contains good states of affairs that include or presuppose these sins.

Notes

221

42 For discussion, see Dunn, Word Biblical Commentary (cited above); see also C.K. Barrett, A Commentary on the Epistle to the Romans (New York: Harper & Row, 1957), 225–28; and C.E.B. Cranfield, Romans 9–16 (London: T&T Clark, 2004), 585–88. 43 For discussion, see Allen P. Ross, Creation & Blessing: A Guide to the Study and Exposition of Genesis (Grand Rapids: Baker Books, 1996), 716. See also Richard J. Clifford and Roland E. Murphy, “Genesis,” in The New Jerome Biblical Commentary, ed. Raymond E. Brown, Joseph A. Fitzmeyer, and Roland E. Murphy (Englewood Cliffs: Prentice Hall, 1990), 40, 43. 44 For discussion, see David G. Peterson, The Acts of the Apostles (Grand Rapids: Eerdmans, 2009), 201. 45 For many more examples from scripture, see Piper, “Are There Two Wills in God?” In Still Sovereign, ed. Thomas R. Schreiner and Bruce A. Ware (Grand Rapids: Baker Books, 2004), 107–31. 46 See, for example, Augustine’s Enchiridion 95, in The Basic Writings of Augustine, Vol. 1, 713; Anselm, Philosophical Fragments, in Anselm of Canterbury, 475–76; and Aquinas, who at least recognizes permission as an “expression” of the divine will at ST 1.19.12. 47 See ST 1.22.2 ad 2, ST 1.48.2 ad 3, and SCG 3.71.6. 48 ST 1.19.9. 49 If it is possible for a human act to be temporally subsequent but explanatory prior to a divine act—for example, if it is possible for God to answer a prayer made at some time in the future relative to what God brings about to answer that prayer—then it may be that two acts could be both antecedent and consequent to one another, in different senses of antecedent and consequent. For present purposes, I need not develop these nuances. 50 None of these points entail that God is the cause of the moral standard, as discussed in Section 6.4. 51 For objections along these lines, see three articles by Katherin A. Rogers, “Does God Cause Sin? Anselm of Canterbury versus Jonathan Edwards on Human Freedom and Divine Sovereignty,” Faith and Philosophy 20 (2003): 371–78; “God Is Not the Author of Sin: An Anselmian Response to McCann,” Faith and Philosophy 24 (2007): 300–10; and “Anselm against McCann on God and Sin: Further Discussion,” Faith and Philosophy 28 (2011): 397–415. I will not, of course, be able fully to engage all that Rogers says in these articles but hope to indicate a path for responding to these sorts of objections. 52 I, of course, recognize that agents sometimes will that which they disapprove of in their own right; indeed, as the discussion makes clear, God permissively wills sin, which he disapproves of in its own right. Still, to say that something is contrary to an agent’s will is one way of expressing that the agent disapproves of it in its own right, even if the agent wills that thing in some other sense. For example, a sailor who, in one sense, willfully (i.e., on purpose) jettisons cargo in order to save a sinking ship might explain that jettisoning the cargo, though it had to be done, was against his will, as a way of expressing that he finds the loss of cargo unappealing and contrary to desire. 53 Of course, traditional Molinists were proponents of DUC and thus also thought that God concurred with sinful acts, causing those acts of sin. I quoted Suarez to this effect in Section 1.1. 54 In characterizing God’s degree of involvement in sin on the two versions of Open Theism, I am assuming what I argued in the previous sections is necessary for these accounts to respond to the problem of moral evil.

222

Notes

55 John H. Hayes, Amos: The Eighth Century Prophet (Nashville: Abingdon Press, 1988), 125–26. 56 See Piper, “Are There Two Wills in God?” 119. See Piper’s essay, also, for additional examples of God’s involvement in sin. 57 Eichrodt goes so far as to say that it is a “theological dilution” to explain such passages in terms of God’s “permissive will.” In his view, mere permission doesn’t quite capture such passages’ portrayal of the level of God’s activity in relation to sin. On the contrary, “in the cases mentioned what is involved is a real act of God.” See Eichrodt, Theology of the Old Testament, Vol. II, 178. Eichrodt gives other examples. Many theologians and biblical commentators, to be sure, have attempted to sanitize such passages in various ways to distance God as much as possible from involvement in the occurrence of sin and evil. Such authors (like all of us) are influenced by moral and metaphysical assumptions they bring to the text, assumptions that, I suspect, often guide their judgments regarding what is needed to make these passages consistent with passages that speak of God’s justice and righteousness. The interpretive questions are sufficiently complicated that due modesty cautions against ruling out completely such sanitized readings. Nevertheless, I must confess that these readings generally seem a stretch to me and less plausible than more straightforward readings that recognize a level of God’s involvement in the occurrence of sin and evil that we may initially find surprising and, perhaps, unsettling. It is worth considering, then, an account that tries to accommodate a more active level of involvement in the occurrence of sin alongside the divine innocence that all parties (myself included) rightfully want to preserve. 58 For a recent extended argument that God, while morally perfect, is not bound by the same ethics or moral norms as human beings, see Mark C. Murphy, God’s Own Ethics: Norms of Divine Agency and the Argument from Evil (Oxford: Oxford University Press, 2017).

Chapter 8 1

2 3

Notice that the foregoing problem does not arise for God’s knowledge of necessary truths, truths that God knows in all possible worlds. For, if the truth that God knows is necessary, then the entity intrinsic to God required by (KT) for God’s knowledge of that truth could be identical to God. Of course, further questions arise about how God could know a multiplicity of necessary truths without introducing a multiplicity of entities intrinsic to God. For more on this, see Aquinas, Summa theologiae 1.15.2. For more on what I will call the “belief ” and “immediate cognition” models, see my “Divine Simplicity, Contingent Truths, and Extrinsic Models of Divine Knowing,” Faith and Philosophy 29, no. 3 (2012): 254–74. Sense-datum theories of perception, for instance, view perceiving as a relation to a sense-datum, while disjunctive theories of perception view perceiving as a relation to an ordinary physical object. See Tim Crane, “The Problem of Perception,” in The Stanford Encyclopedia of Philosophy (Spring 2017 Edition), ed. E.N. Zalta http:// plato.standford.edu/archives/spr2017/entries/perception-problem/. See also Joseph Campbell, Reference and Consciousness (Oxford: Oxford University Press, 2002), 114–20; and Tom Stoneham, “A Neglected Account of Perception,” Dialectica 62 (2008): 307–22. Stoneham (313) even suggests that perceptions are relations to

Notes

223

objects that might come and go without any intrinsic change in the perceiving subject. 4 Michael Thau, Consciousness and Cognition (Oxford: Oxford University Press, 2002), 62. 5 Thau, Consciousness and Cognition, 66. Thau defends the view again in “Precis,” Philosophical Studies 132 (2007): 565–70. See also Timothy Williamson, Knowledge and Its Limits (Oxford: Oxford University Press, 2000), who (21) takes both “believing” and “knowing” to be relations of subjects to propositions. 6 See, for example, William Lane Craig, God over All. 7 William P. Alston, “Does God Have Beliefs?” Religious Studies 22 (1986): 297–98. See also Morris, Our Idea of God, 84 and Rogers, Perfect Being Theology, 77. 8 Stoneham (318–20) similarly recommends his “Purely Relational” account of perception for offering a genuinely “direct” account of perception in contrast to standard views on which perceiving takes place by means of an inner representation or experience. 9 Miller, A Most Unlikely God, 97. 10 McCann, Creation and the Sovereignty of God, 48. 11 For a classic statement of this view, see G.E.M. Anscombe, Intention (Oxford: Blackwell, 1957). 12 See, for example, O’Connor, Persons and Causes, 71–72. 13 For an account along these lines, see Brian Shanley, “Aquinas on God’s Causal Knowledge: A Reply to Stump and Kretzmann,” American Catholic Philosophical Quarterly 72, no. 3 (1998): 449. In speaking of what contingent entities “ought” to exist relative to a given subject, I, of course, do not mean to imply that God is bound in justice to bring the thing about, for that would imply that God is failing to do something he ought. 14 That there are any human beings is a contingent fact, but, ultimately, what sets the moral standard for human beings, I assume, is not a contingent fact, and so, in keeping with what was said in Chapter 6, the moral standard is not something that God knows in an act of causing it. 15 William Hasker, “Yes, God Has Beliefs!,” Religious Studies 24, no. 3 (1988): 389. 16 Ibid., 391. 17 I assume that if God can stand in the knowing and/or causal relation to past and future objects and events, then these objects and events can also partially constitute and inform God’s acts of knowing and/or causing them, where these acts are understood as relations to these objects and events. 18 The concerns discussed here have much in common with standard objections to presentism. For a presentist who rejects (NR), see Mark Hinchliff, “A Defense of Presentism” (PhD diss., Princeton University, 1988), chs. 2–3. More commonly, presentists have affirmed (NR), arguing that propositions about the past and future can be true even without relations to past and future events, either because such propositions don’t need truth-makers or because they have truth-makers that exist in the present. For examples of these more common strategies, see Trenton Merricks, Truth and Ontology (Oxford: Oxford University Press, 2007), 119–45; Thomas M. Crisp, “Presentism,” in The Oxford Handbook of Metaphysics, ed. M.J. Loux and D.W. Zimmerman (Oxford: Oxford University Press, 2003), 236–42; and Simon Keller, “Truth and Truthmaking,” Oxford Studies in Metaphysics 1 (2004): 83–104. To the extent that one finds these more common strategies plausible, the foregoing response to the presentist objection loses power.

224

Notes

19 For approaches of this second sort, see Eleonore Stump and Norman Kretzmann, “Eternity,” The Journal of Philosophy 78, no. 8 (1981): 429–58; Brian Leftow, Time and Eternity (Ithaca: Cornell University Press, 1991), 230–35; and Stump, Aquinas, 131–58. 20 As we will see in Section 8.5, however, Dual Sources does offer a new, or at least a neglected, account of how an eternal God can be said to act in time. 21 For a presentation of the Foreknowledge Dilemma, see Linda Zagzebski, “Foreknowledge and Free Will,” in The Stanford Encyclopedia of Philosophy (Summer 2017 Edition), ed. Edward N. Zalta forthcoming URL = http://plato.stanford.edu/ archives/sum2017/entries/free-will-foreknowledge/. 22 Ibid. 23 For an example of this sort of response, see Michael Rota, “The Eternity Solution to the Problem of Human Freedom and Divine Foreknowledge,” European Journal for Philosophy of Religion 2, no. 1 (2010): 165–86. 24 Note that a Dual Sourcer who denied that God’s knowledge of his effects as possibilia is logically prior to the existence of those effects seemingly could not appeal to exemplary causation to interpret the traditional claim “that God’s knowledge is the cause of things.” For an author who has reservations about acknowledging such possibilia prior to creation, see James Ross, “Aquinas’s Exemplarism, Aquinas’s Voluntarism,” American Catholic Philosophical Quarterly, 64, no. 2 (1990): 171–98. 25 Strictly speaking, a Dual Sourcer could adopt the belief or immediate cognition model and, on the assumption that all points in time are present to God, give an alternative account of God’s knowledge of what free creatures will freely do. Because I prefer the agency model, I have developed an approach in terms of that model. 26 Although I have not used the term “strict dependence” before now, the basic distinction and contrast with causal or explanatory dependence have been operative since Chapter 4. For explicit use of this terminology, together with a fuller discussion of divine aseity, see my “Aseity,” in Routledge Encyclopedia of Philosophy (2015): http://www.rep.routledge.com/articles/thematic/aseity/v-1., doi: 10.4324/9780415249126-K3570-1. 27 Richard Rice, “Biblical Support for a New Perspective,” in The Openness of God (Downers Grove, IL: InterVarsity Press, 1994), 12. 28 Ibid., 11. 29 Clark H. Pinnock, “Systematic Theology,” in The Openness of God, 113. 30 Ibid., 113, 116. 31 Thus, William Hasker once wrote, in a way that still represents the dominant view, that “if you wish to maintain a strong doctrine of providence.… and yet to uphold libertarian free will, then Molinism is the only game in town.” See Hasker’s “Providence and Evil,” 97. For a more recent statement of this common assumption, see Derk Pereboom, “Libertarianism and Theological Determinism,” 129. 32 For a clear presentation and defense of the Molinist approach, see Flint, Divine Providence: The Molinist Account. 33 For a classic statement of this “grounding objection” to middle knowledge, see Robert Merrihew Adams, “Middle Knowledge and the Problem of Evil,” American Philosophical Quarterly 14, no. 2 (1977): 109–17. For a more recent statement, see Pereboom, “Libertarianisim and Theological Determinism,” where (114) Pereboom writes: “But Molinism is a highly controversial position, not least because it is not clear how there could be truths about what non-actual free creatures would freely decide on which God could base his decisions as to which to actualize. Truths about

Notes

34 35 36

37

38

39 40 41

225

what creatures would freely decide would presumably be grounded in what they in fact freely decide, or at least in what they will freely decide, but if they don’t exist and never will, such grounding is unavailable.” I, of course, do not claim in these few paragraphs to have demonstrated that Molinism is false. Any attempt to do so would require an in-depth engagement with Molinist responses to these objections. See, for example, Aquinas, ST 1.25.3, and Ott, Fundamentals of Catholic Dogma, 47. For evidence of this standard teaching, see Timothy Pawl, “Free Will and Grace,” in The Routledge Companion to Free Will, ed. Kevin Timpe, Meghan Griffith, and Neil Levy (New York: Routledge, 2017), 531–42; and Ott, Fundamentals of Catholic Dogma, 229–36. As Ott makes clear, according to Catholic theology, human beings can perform some good acts without grace after the Fall, but not good acts of the sort that can lead to salvation. The need for God’s grace has been attributed principally to two reasons. First, the Fall so wounded and weakened human nature that we are no longer able to will and do the good without the remedial help of God’s grace. Second, even apart from the Fall, the actions needed for salvation and eternal friendship with God surpass what is possible to our natural human capacities; thus, grace is needed to elevate our capacities so that we can perform supernatural acts of friendship with God of a sort that go beyond what is proper or possible for human nature considered apart from grace. These two reasons are not mutually exclusive. Catholics have emphasized both, whereas some Protestants emphasize only the first. Pawl (536–38) gives evidence of both traditional and contemporary support for the claim that God gives to everyone the preceding grace requisite to make good or salutary acts possible. Notice that a bad act (or refraining from a good act) could satisfy the “broad” conception of what it is for an act to be free in the libertarian sense (introduced in Section 1.2), even if immediately preceding the act God had not provided the grace requisite to make that act possible. In such a circumstance, the creature’s antecedent state would, in fact, determine the creature not to perform a good act. But if this antecedent state, with its lack of grace, were the result of some prior bad choice made under conditions when God had given the grace needed to perform a good act, then the creature would be responsible for this determining antecedent state. But, then, the act determined by this antecedent state would be free in the “broad” libertarian sense. The upshot: given broad libertarianism, a creature’s bad acts could be free in the libertarian sense, even if because of prior faults many of these bad acts are performed absent the grace that would have been needed for the creature to perform a good act. For the sake of simplicity, I will, in what follows, typically be assuming strict libertarianism. Saint Augustine, Grace and Free Will, 20, trans. Robert P. Russell, in The Fathers of the Church: A New Translation, Vol. 59 (Washington, DC: The Catholic University of America Press, 1968), 271.The translations of scripture are from the Augustine, but I will identify the scriptural references parenthetically, as above. Ibid., ch. 21, 273. Ibid., ch. 32, 287. Ibid., ch. 31, 284–85. In this connection, Augustine asks, “Why do we pray that those who refuse to believe should come to believe? It would be absolutely useless for us to do this unless we were entirely correct in our belief that Almighty God can bring about a conversion of the will to believe even in those whose wills are perverse and hostile to the faith. The Psalmist does in fact put his finger on man’s free will where he says: ‘Today if you shall hear His voice, harden not your hearts’ (Ps. 95:8). But if

226

42 43 44 45 46 47 48 49

50 51

52

53 54

Notes God were not able to take away even that hardness of heart, He could not declare through his Prophet: ‘I will take away the stony heart out of their flesh, and will give them a heart of flesh’” (Ezek 11:19). Ibid., ch. 29, 282. Ibid., ch. 15, 267. Ibid., ch. 19, 270. Ibid., ch. 10, 261–62. Ibid., ch. 19, 270. Elsewhere he cites Jesus’s words at Matthew 16:27, supporting the claim that we will be repaid for our works. Ibid., ch. 10, 262. Ibid., ch. 19, 271. Ibid., ch. 20, 272. Ibid., ch. 15, 267. As should be evident, Augustine reads Paul’s teaching that grace rendered for works is not really grace as applying only to works that are not themselves gifts of grace. Such a reading is needed to render these claims consistent with other things Paul says, as well as claims made elsewhere in scripture. See ibid., 20, 271: “How, then, are we to understand [Paul’s] expression, ‘not as the outcome of works, lest anyone may boast’?.… These refer to works that come from yourself, as your own, and not to those in which God has molded you, namely, works in which God has formed and created you.” Indeed, this conception would appear to be affirmed in Catholic doctrine. See, for example, chapter 16 and Canon 32 of the Council of Trent’s Decree on Justification. See also, the Second Council of Orange, especially canons 6, 9, 20, and 25. There is, of course, no reason that one and the same grace can’t be, for example, both consequent and concurrent (etc.), relative to different creaturely actions. For example, a concurrent grace in which God brings about a good work is concurrent relative to that work, but consequent relative to a prior act of prayer that God would grant that good work. Strictly speaking, antecedent grace could determine a good act and be consistent with the “broad” conception of what it is for an act to be free in the libertarian sense, provided that the antecedent grace in question were given as a reward for some previous good act that was performed in the context of antecedent grace that did not determine that previous good act. Augustine is very clear, for example, that God can give good acts to a positively stony and ill-bent heart. See Grace and Free Will, ch. 12, 263, and ch. 29, 282. See, also, Aquinas ST 2–2.82.3 and ST 1–2.112.2. The problems with the first and second rebuttals are similar to some of the ones discussed by Taylor W. Cyr and Matthew T. Flummer, “Free Will, Grace, and AntiPelagianism,” International Journal for Philosophy of Religion 83, no. 2 (2018): 183–99; by Simon Kittle, “Grace and Free Will: Quiescence and Control,” Journal of Analytic Theology 3 (2015): 89–108; and by C.P. Ragland, “The Trouble with Quiescence: Stump on Grace and Freedom,” Philosophia Christi 8, no. 2 (2006): 343–62. These authors criticize proposals that make the resistance or nonresistance to grace up to the creature alone, on the grounds that it makes the creature praiseworthy for something apart from grace, namely, praiseworthy for not resisting the grace. Notice that the first objection to the second rebuttal could be blocked were it denied that the nonresistance is good and meritorious. But, even if we allowed (for the sake of argument) this controversial denial, the second and third problems with the second rebuttal would remain. For the proposals discussed by the aforementioned authors, see Richard Cross, “Anti-Pelagianism and the Resistibility of Grace,” Faith and

Notes

55

56

57 58

227

Philosophy 22 (2005): 199–210; Eleonore Stump, Aquinas (London: Routledge, 2003), 389–404; and Kevin Timpe, “Grace and Controlling What We Do Not Cause,” Faith and Philosophy 24 (2007): 284–99. The focus has been on meritorious acts, but what about meritorious refrainings? Are these God’s gifts on Dual Sources? And are they ultimately up to God as well as the creature? First, many refrainings involve a positive choice to refrain from some act; thus, they involve a meritorious act—the choice to refrain—that God gives in an act of concurrent grace, as described above. If there are cases of meritorious refraining that do not involve any positive choice to refrain, then even these too are ultimately up to God, and hence are God’s gifts, on Dual Sources—for, on Dual Sources, God will have given antecedent graces that make possible and incline me toward the meritorious refraining. Moreover, on Dual Sources, if I don’t perform some act A in accordance with the antecedent grace inclining me to refrain, it follows that God didn’t cause my performing A when he could have. And, thus, just as it is ultimately up to me that I not perform A in the context of the antecedent grace, so it is also ultimately up to God. And, so, if it is a merit that I not perform A, that merit is a gift from God. It is doubtful that Molinism provides an alternative. For, putting aside the problem that the truth of the counterfactuals of freedom cannot be ultimately up to creatures that do not exist, even if they could be, the Molinist situation would be one where it is up to the creature alone whether, given the antecedent grace bestowed by God, it is true that the creature would perform the good act or that the creature would not perform it. And so, unlike Dual Sources, Molinism cannot do justice to the claim that our good acts are through and through God’s gifts. See Saint Augustine, Admonition and Grace, trans. John Courtney Murray, in The Fathers of the Church: A New Translation, Vol. 2 (Washington, DC: The Catholic University of America Press, 1950), ch. 38, 292. For newcomers to the literature on grace, the sheer multiplicity of names for different types and classifications of grace can be daunting. I have distinguished three types— antecedent, concurrent, and consequent—with most of the discussion focused on the first two. For the sake of making connections with the wider literature and tradition, what I call antecedent grace is similar to, or a version of, what is sometimes called prevenient (praeveniens) or operating (operans) grace, while concurrent grace is similar to, or a version of, what is sometimes called concomitant (concomitans) or cooperating (co-operans) grace. Details vary depending on the account. For example, my concurrent grace has in common with most accounts of cooperating grace that, in contrast to antecedent or prevenient grace, it coincides with, rather than precedes, the good creaturely act. But other accounts of cooperating grace do not build around or exploit the extrinsic model of divine agency, the way my account of concurrent grace does. To mention two other types of grace: traditionally, sufficient grace (gratia sufficiens) denotes grace that confers on the creature an ability to perform a good act, but which is consistent with the good act’s not being performed, while efficacious grace (gratia efficax) ensures or guarantees the performance of the good act. Notice that as “sufficient” is used in discussions of sufficient grace, it does not mean “logically sufficient” as I have been using that term in this work, in keeping with usage in contemporary freewill debates. Sufficient grace is consistent with the nonperformance of the good act, but grace that is “logically sufficient” for a good act is not consistent with that act’s nonperformance. Sufficient grace, therefore, corresponds most closely to my antecedent grace, while efficacious grace, which guarantees the good act, corresponds most closely

228

Notes

to my concurrent grace, since concurrent grace is logically sufficient for the act. Some accounts of efficacious grace would understand that grace to be antecedent to the good act. But, for reasons that should now be clear, efficacious antecedent grace would not be consistent with the act’s being free in the libertarian sense, since it would introduce a prior and logically sufficient condition for that act. For a concise overview of different types of grace, see Ott, Fundamentals of Catholic Dogma, 220–22. 59 See Alston, Divine Nature and Human Language, 148–49. 60 Ibid., 148. 61 See, for example, Stump, Aquinas, 150–51. 62 Alston, Divine Nature and Human Language, 148. 63 Ibid., 152, 158. 64 Ibid., 157. 65 Ibid., 158. 66 Herbert McCabe, God Matters, 221–22. 67 See Heinrich Denzinger, Compendium of Creeds, Definitions, and Declarations on Matters of Faith and Morals, 43rd edition, ed. Peter Hunermann (San Francisco: Ignatius Press, 2012), 135. The quotation is taken from the third canon of the council, the relevant portion of which reads in Latin “Si quis invocation humana gratiam Dei dicit posse conferri, non autem ipsam gratiam facere, ut invocetur a nobis, contradicit Isaiae prophetae vel Apostolo.… ” The translation uses “prompts” for “facere,” but “facere” could and perhaps should be translated as “makes.” 68 Augustine, On the Gift of Perseverance, 64. In Augustine, Four Anti-Pelagian Writings, trans. John A. Mourant and William J. Collinge (Washington, DC: The Catholic University of America Press, 1992), 333. 69 Aquinas, ST 2–2.83.16 ad 1. 70 Karl Barth, Prayer, second edition trans. Sara F. Terrien (Philadelphia: The Westminster Press, 1985), 36. 71 See Augustine, Two Books on Genesis against the Manichees, 1. 22.34, trans. Roland J. Teske (Washington, DC: The Catholic University of America Press, 1991), 82. See also, Augustine, On the Gift of Perseverance, 64. 72 Paul K. Moser, “Prayer.” From an entry entitled “Prayer,” at http://pmoser.sites.luc. edu/idolanon/Prayer.html. Accessed June 6, 2018. 73 Joseph A. Fitzmeyer, “The Letter to the Romans,” in The New Jerome Biblical Commentary, ed. Raymond E. Brown, Joseph A. Fitzmeyer, and Roland E. Murphy, (Englewood Cliffs, NJ: Prentice Hall, 1990), 855. 74 St. Teresa of Avila, “In the Hands of God,” in The Collected Works of St. Teresa of Avila, Vol. Three, trans. Kieran Kavanaugh, and Otilio Rodriguez (Washington, DC: Institute of Carmelite Studies, 1985), 378. 75 Julian of Norwich, Showings, trans. Edmund Colledge, and James Walsh (New York: Paulist Press, 1978), 253. 76 Ibid., 254. 77 The Liturgy of the Hours, Vol. 1 (New York: Catholic Book Publishing Co., 1975), 41. The invitatory is inspired by Psalm 51, a psalm that in general supports a robust conception of God’s grace. 78 The Catechism of the Catholic Church (United States Catholic Conference, Inc.— Libreria Editrice Vaticana, 1994), # 2564. 79 Are some prayers defective? Do some instances of human dialogue with God consist in yelling a string of blasphemies at him? Dual Sources can accommodate affirmative answers to both questions. Yelling a string of blasphemies would constitute a sinful

Notes

80 81 82

83

84

85

86 87 88 89 90

229

act and can be handled along the lines suggested for other sins in Chapters 6 and 7. Some defective prayers are arguably still good acts that bring us into friendship with God. But their defects can be analyzed as privations, also along the lines suggested in Chapter 6. M.J. Langford, “Predestination,” in The Westminster Dictionary of Christian Theology, ed. Alan Richardson and John Bowden (Philadelphia: The Westminster Press, 1983), 460. Joseph Pohle, “Predestination,” in The Catholic Encyclopedia (New York: Robert Appleton Company, 1911). Retrieved May 24, 2017, from New Advent http://www. newadvent.org/cathen/1237a.htm See, for example, Saint Augustine, On the Predestination of the Saints and On the Gift of Perseverance. In Augustine, Four Anti-Pelagian Writings; Aquinas, ST 1.23; and John Calvin, Institutes of the Christian Religion, trans. Henry Beveridge (Grand Rapids, MI: Eerdmans, 1989), Bk. 3, chs. 21–24. I do not, of course, claim that these thinkers have exactly the same doctrine of predestination, in all details; nor that any of them affirmed the precise wording of Predestination; nor that Predestination encompasses all that each of them believes on the topic. But I do think that Predestination represents a common core that would be affirmed by these and other thinkers in the tradition. See, for example, Rom. 8:29–30, Rom. 9, Acts 13:48, and Jn 10:25–29. For a recent discussion of the biblical roots of the doctrine of predestination, as well as the controversy surrounding the interpretation of such passages, see especially chapter 1 of Matthew Levering, Predestination: Biblical and Theological Paths (Oxford: Oxford University Press, 2011). The prayer is from Eucharistic Prayer I, see the Daughters of St. Paul, Saint Paul Daily Missal (Boston: Pauline Books & Media, 2012), 849. See, also, from the same Mass (847): “Remember Lord, your servants N and N and all gathered here.… For them, we offer you this sacrifice of praise or they offer it for themselves and all who are dear to them: for the redemption of their souls.… ” Similar prayers are commonly found in the Mass in the prayers over the offering and after communion. See, for example (586): “Look with kindness upon your people, O Lord, and grant, we pray, that those you were pleased to renew by eternal mysteries may attain in their flesh the incorruptible glory of the resurrection.” For other instances, see 596 and 646. The Liturgy of the Hours, Vol. 4 (New York: Catholic Book Publishing Co., 1975), 970–71. Such prayers abound in the Liturgy of the Hours. For another example (1227): “Lord, … may we live our days in quiet joy and, with the help of the Virgin Mary’s prayers, safely reach your kingdom. Grant this through Christ our Lord.” Interestingly, right at the start of Rom. 10, following texts in Chapters 8 and 9 that many have read as supporting Predestination, St. Paul writes with the Israelites in mind: “My heart’s desire and prayer to God for them is that they be saved.” See, for example, Mt. 19:16–17, Mt. 16:27, Rom. 2:6–8, Jn 3:14–18, Acts 16:31, and Acts 3:19. See, for example, Augustine, Grace and Free Will, ch. 19, 270, and Aquinas ST 1–2.113.1–5. Oliver Crisp, “Is Universalism a Problem for Particularists?” Scottish Journal of Theology 63, no. 1 (2010): 7. For some discussion of this Molinist approach, see Ott, Fundamentals of Catholic Dogma, 243–44.

230

Notes

91 For example, Molinism might enable a reconciliation of a libertarian reading of Responsibility with just the first conjunct of Predestination, denying the second conjunct, which holds that “for any possible created person, it is within God’s power (whether or not) to choose that that person attain salvation.” 92 Strictly speaking, a Dual Sourcer need not accept the stipulation that God gives the offer of salvation and the necessary antecedent graces to all created persons. If, however, a Dual Sourcer were to reject this stipulation, and assuming that the antecedent offer and graces are necessary for a person to attain salvation, then he would need to say that God makes two choices that a person attain salvation. The first would consist in God’s giving the antecedent offer and graces in order that the person attain salvation. The second would consist in God’s bringing about the person’s accepting the offer. The first would be prior to, but not logically sufficient for, the person’s accepting. The second would be logically sufficient, but not prior. The account is somewhat simpler given the stipulation. 93 For Augustine, see On the Gift of Perseverance, 28. For Aquinas, see ST 1.23.5 ad 3. Crisp, “Is Universalism a Problem for Particularists?” esp. 22–23; John Lamont, “The Justice and Goodness of Hell,” Faith and Philosophy 28, no. 2 (2011): esp. 171–73; Paul A. Macdonald Jr., “Hell, the Problem of Evil, and the Perfection of the Universe,” American Catholic Philosophical Quarterly 89, no. 4 (2015): 603–28. Macdonald, I think rightly, argues that while a world in which those who do not accept the offer of salvation would be a good world in reflecting God’s justice in a certain way, God is not bound to create such a world. A world in which all attain salvation would also be a good world, and Macdonald contends that creating either sort of world would be consistent with God’s justice and goodness. 94 Crisp, “Presentism,” 22. It is not entirely clear whether Crisp affirms this suggestion, but, at the very least, he takes it seriously. 95 Rom. 9:14, 21–23. I, of course, am not claiming that this question is the only issue taken up in Rom. 9. 96 See 1 Tim. 2:4–5 and 2 Pet. 3:9. 97 For an interpretation of 1 Tim. 2:4 along these lines, see Aquinas, ST 1.23.4 ad 3. See also, Long, “St. Thomas Aquinas, Divine Causality, and the Mystery of Predestination”, esp. 73–76. 98 See, for example, Mt. 7:13, Mt. 25:46, 1 Cor. 1:18, and Mt. 26:24. 99 A Dual Sourcer who rejects (d) will presumably reject (c) rather than (a) or (b), because we are assuming she is committed to (a) in virtue of being a Dual Sourcer and because it would seem odd to reject (d) without thinking that (b) is true.

Bibliography Adams, Marilyn McCord. Horrendous Evils and the Goodness of God. Ithaca, NY: Cornell University Press, 1999. Adams, Robert Merrihew. “Middle Knowledge and the Problem of Evil.” American Philosophical Quarterly 14, no. 2 (1977): 109–17. Aertsen, Jan. Nature and Creature: Thomas Aquinas’s Way of Thought. Leiden: E.J. Brill, 1988. Alexander, David E. Goodness, God, and Evil. New York: Continuum, 2012. Alston, William P. “Does God Have Beliefs?” Religious Studies 22, no. 3/4 (1986): 287–306. Alston, William P. Divine Nature and Human Language. Ithaca, NY: Cornell University Press, 1989. Anglin, Bill, and Stewart Goetz. “Evil Is Privation.” International Journal for Philosophy of Religion 13, no. 1 (1982): 3–12. Anscombe, G.E.M. Intention. Oxford: Blackwell, 1957. Anscombe, G.E.M. “Causality and Determination.” In Metaphysics and the Philosophy of Mind: The Collected Papers of G.E.M. Anscombe, Vol. 2. Minneapolis: University of Minnesota Press, 1981, 133–47. Anselm. De concordia. Translated by Thomas Bermingham. In Davies and Evans, eds., Anselm of Canterbury, 435–74. Anselm. Monologion. Translated by Simon Harrison. In Davies and Evans, eds., Anselm of Canterbury, 5–81. Anselm. On the Fall of the Devil. Translated by Ralph McInerny. In Davies and Evans, eds., Anselm of Canterbury, 193–232. Anselm. Philosophical Fragments. Translated by D.P. Henry. In Davies and Evans, eds., Anselm of Canterbury, 475–501. Anselm. Proslogion. Translated by M.J. Charlesworth. In Davies and Evans, eds., Anselm of Canterbury, 82–104. Aquinas. Quastiones disputatae de potentia dei. Translated under the title On the Power of God by the Fathers of the English Dominican Province. London: Burns Oates & Washbourne Ltd, 1932. Aquinas. Summa contra gentiles. Translated by James F. Anderson, Anton C. Pegis, Vernon J. Bourke, and Charles J. O’Neil. Notre Dame: University of Notre Dame Press, 1975. Aquinas. Summa theologica. Translated by the Fathers of the English Dominican Province. Westminster, MD: Christian Classics, 1981. Aquinas. Quodlibetal Questions 1 and 2. Translated by Sandra Edwards. Toronto: Pontifical Institute of Medieval Studies, 1983. Aquinas. Commentary on Aristotle’s De Interpretatione, 1.9. In Aquinas: Selected Philosophical Writings. Selected and translated by Timothy McDermott. Oxford: Oxford University Press, 1993, 277–84. Aquinas. De ente et essentia. In Aquinas: Selected Philosophical Writings. Selected and translated by Timothy McDermott. Oxford: Oxford University Press, 1993, 90–113. Aquinas. Commentary on Aristotle’s Metaphysics. Translated by John P. Rowan. Notre Dame: Dumb Ox Books, 1995.

232

Bibliography

Aquinas. De malo. Translated under the title On Evil by Jean Oesterle. Notre Dame: University of Notre Dame Press, 1995. Aquinas. Commentary on the Book of Causes. Translated by Vincent A. Guagliardo, Charles R. Hess, and Richard Taylor. Washington DC: The Catholic University of America Press, 1996. Aquinas. Commentary on Aristotle’s Physics. Translated by Richard J. Blackwell, Richard J. Spath, and W. Edmund Thirlkel. Notre Dame: Dumb Ox Books, 1999. Aquinas. An Exposition of the “On the Hebdomads” of Boethius. Translated by Janice L. Schultz and Edward A. Synan. Washington, DC: The Catholic University of America Press, 2001. Aristotle. Metaphysics. Translated by W.D. Ross. In The Basic Works of Aristotle. Edited by Richard McKeon. New York: Random House, 1941, 681–926. Aristotle. Physics. Translated by R.P. Hardie and R.K. Gaye. In The Basic Works of Aristotle. Edited by Richard McKeon. New York: Random House, 1941, 213–394. Augustine. The Enchiridion on Faith, Hope, and Love. Translated by J.F. Shaw. In Basic Writings of Saint Augustine, Vol. 1. Edited by Whitney J. Oates. New York: Random House Publishers, 1948. Augustine. Admonition and Grace. Translated by John Courtney Murray. In The Fathers of the Church: A New Translation, Vol. 2. 2nd ed. Washington, DC: The Catholic University of America Press, 1950, 245–305. Augustine. Grace and Free Will. Translated by Robert P. Russell. In The Fathers of the Church: A New Translation, Vol. 59. Washington, DC: The Catholic University of America Press, 1968, 243–308. Augustine. Two Books on Genesis against the Manichees. Translated by Roland J. Teske. In The Fathers of the Church: A New Translation, Vol. 84. Washington, DC: The Catholic University of America Press, 1991, 45–141. Augustine. On the Gift of Perseverance. Translated by John A. Mourant and William J. Collinge. In Four Anti-Pelagian Writings, from The Fathers of the Church: A New Translation, Vol. 86. Washington, DC: The Catholic University of America Press, 1992, 271–337. Augustine. On the Predestination of the Saints. Translated by John A. Mourant and William J. Collinge. In Four Anti-Pelagian Writings, from The Fathers of the Church: A New Translation, Vol. 86. Washington, DC: The Catholic University of America Press, 1992, 218–70. Augustine. The City of God against the Pagans. Translated by R.W. Dawson. Cambridge: Cambridge University Press, 1998. Baker, Lynne Rudder. “Why Christians Should Not Be Libertarians: An Augustinian Challenge.” Faith and Philosophy 20, no. 4 (2003): 460–78. Barclay, John M.G. “Introduction.” In Divine and Human Agency in Paul and His Cultural Environment. Edited by John M.G. Barclay and Simon J. Gathercole. New York: T&T Clark, 2006, 1–8. Barrett, C.K. A Commentary on the Epistle to the Romans. New York: Harper & Row, 1957. Barth, Karl. Prayer. 2nd ed. Translated by Sara F. Terrien. Philadelphia: The Westminster Press, 1985. Basinger, David. “Practical Implications.” In Pinnock and others, eds., The Openness of God, 155–76. Bennett, Karen. “Why the Exclusion Problem Seems Intractable, and How, Just Maybe, to Tract It.” Nous 37, no. 3 (2003): 471–97.

Bibliography

233

Bergmann, Michael, and Jeffrey E. Brower. “A Theistic Argument against Platonism and in Support of Truthmakers and Divine Simplicity.” Oxford Studies in Metaphysics 2 (2006): 357–86. Bontly, Thomas D. “Exclusion, Overdetermination, and the Nature of Causation.” Journal of Philosophical Research 30 (2005): 261–82. Boyd, Gregory A. God of the Possible: A Biblical Introduction to the Open View of God. Grand Rapids, MI: Baker Books, 2000. Boyle, Joseph, Jr., Germain Grisez, and Olaf Tollefsen. Free Choice: A Self-Referential Argument. Notre Dame: University of Notre Dame Press, 1976. Brock, Stephen L. Action and Conduct: Thomas Aquinas and the Theory of Action. Edinburgh: T&T Clark, 1998. Brower, Jeffrey E. “Simplicity and Aseity.” In The Oxford Handbook of Philosophical Theology. Edited by Thomas P. Flint and Michael Rea. Oxford: Oxford University Press, 2009, 105–28. Brower, Jeffrey E. “Medieval Theories of Relations.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Winter 2015 Edition, https://plato.stanford. edu/archives/win2015/entries/relations-medieval/. Burrell, David. Freedom and Creation in Three Traditions. Notre Dame: University of Notre Dame Press, 1993. Calder, Todd C. “Is the Privation Theory of Evil Dead?” American Philosophical Quarterly 44, no. 4 (2007): 371–81. Campbell, Joseph. Reference and Consciousness. Oxford: Oxford University Press, 2002. Carey, Brandon. “Overdetermination and the Exclusion Problem.” Australasian Journal of Philosophy 89, no. 2 (2011): 251–62. Catechism of the Catholic Church. United States Catholic Conference, Inc.—Libreria Editrice Vaticana, 1994. Chisholm, Roderick M. On Metaphysics. Minneapolis: University of Minnesota Press, 1989. Clarke, Randolph. Libertarian Accounts of Free Will. Oxford: Oxford University Press, 2003. Clarke, Randolph. “Libertarian Views: A Critical Survey of Noncausal and Event-Causal Accounts of Free Agency.” In Robert Kane, ed., The Oxford Handbook of Free Will, 356–85. Clarke, W. Norris. “The Meaning of Participation in St. Thomas.” Proceedings of the American Catholic Philosophical Association 26 (1952): 147–59. Clarke, W. Norris. Explorations in Metaphysics. Notre Dame: University of Notre Dame Press, 1994. Clarke, W. Norris. The One and the Many: A Contemporary Thomistic Metaphysics. Notre Dame: University of Notre Dame Press, 2001. Clifford, Richard J., and Roland E. Murphy “Genesis.” In The New Jerome Biblical Commentary. Edited by Raymond E. Brown, Joseph A. Fitzmeyer, and Roland E. Murphy. Englewood Cliffs, NJ: Prentice Hall, 1990, 8–43. Copan, Paul, and William Lane Craig. Creation out of Nothing. Grand Rapids, MI: Baker Academic, 2004. Craig, William Lane. The Only Wise God. Grand Rapids, MI: Baker Book House Company, 1987. Craig, William Lane. “Creation, Providence, and Miracles.” In Philosophy of Religion: A Guide to the Subject. Edited by Brian Davies. Washington, DC: Georgetown University Press, 1998, 136–62.

234

Bibliography

Craig, William Lane. God over All: Divine Aseity and the Challenge of Platonism. Oxford: Oxford University Press, 2016. Crane, Tim. “The Problem of Perception.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Spring 2017 Edition, https://plato.stanford.edu/archives/ spr2017/entries/perception-problem/. Cranfield, C.E.B. Romans 9–16. London: T&T Clark, 2004. Crisp, Oliver. “Is Universalism a Problem for Particularists?” Scottish Journal of Theology 63, no. 1 (2010): 1–23. Crisp, Thomas M. “Presentism.” In The Oxford Handbook of Metaphysics. Edited by M.J. Loux and D.W. Zimmerman. Oxford: Oxford University Press, 2003, 211–45. Crosby, John F. “Is All Evil Only Privation?” Proceedings of the American Catholic Philosophical Association 75 (2002): 197–209. Crosby, John F. “Doubts about the Privation Theory That Will Not Go Away: A Response to Patrick Lee.” American Catholic Philosophical Quarterly 81, no. 3 (2007): 489–505. Cross, Richard. “Anti-Pelagianism and the Resistibility of Grace.” Faith and Philosophy 22, no. 2 (2005): 199–210. Curley, Edwin. “The Incoherence of Christian Theism.” The Harvard Review of Philosophy 11, no. 1 (2003): 74–100. Cyr, Taylor W., and Matthew T. Flummer. “Free Will, Grace, and Anti-Pelagianism.” International Journal for Philosophy of Religion 83, no. 2 (2018): 183–99. Daughters of St. Paul. Saint Paul Daily Missal. Boston: Pauline Books & Media, 2012. Davies, Brian. The Thought of Thomas Aquinas. Oxford: Oxford University Press, 1992. Davies, Brian. An Introduction to the Philosophy of Religion. 2nd ed. Oxford: Oxford University Press, 1993. Davies, Brian. The Reality of God and the Problem of Evil. New York: Continuum, 2006. Davies, Brian. Thomas Aquinas on God and Evil. Oxford: Oxford University Press, 2011. Davies, Brian, and G.R. Evans, eds. Anselm of Canterbury: The Major Works. Oxford: Oxford University Press, 1998. Dentan, R.C. “Heart.” In The Interpreter’s Dictionary of the Bible. Edited by George Arthur Buttrick and others. Nashville: Abingdon Press, 1962, 549–50. Denzinger, Heinrich. Compendium of Creeds, Definitions, and Declarations on Matters of Faith and Morals. 43rd ed. Edited by Peter Hunermann. San Francisco, CA: Ignatius Press, 2012. Descartes, René. Meditations on First Philosophy. 3rd ed. Translated by Donald A. Cress. Indianapolis, IN: Hackett, 1993. Dodds, Michael J. Unlocking Divine Action: Contemporary Science & Thomas Aquinas. Washington, DC: The Catholic University of America Press, 2012. Dunn, James D.G. Word Biblical Commentary, Vol. 38b, Romans 9–16. Dallas: Word Book Publishers, 1988. Edwards, Douglas R., and Mark Allan Powell. “Heart.” In The HarperCollins Bible Dictionary. 3rd ed. Edited by Mark Allan Powell and others. New York: HarperOne, 2011, 368. Eichrodt, Walther. Theology of the Old Testament, Vol. 2. Translated by J.A. Baker. Philadelphia: The Westminster Press, 1967. Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Heusenstamm: Editiones Scholasticae, 2014. Feser, Edward. Five Proofs of the Existence of God. San Francisco, CA: Ignatius Press, 2017.

Bibliography

235

Fitzmeyer, Joseph A. “The Letter to the Romans.” In The New Jerome Biblical Commentary. Edited by Raymond E. Brown, Joseph A. Fitzmeyer, and Roland E. Murphy. Englewood Cliffs, NJ: Prentice Hall, 1990, 830–68. Flint, Thomas P. “Compatibilism and the Argument from Unavoidability.” The Journal of Philosophy 84, no. 8 (1987): 423–40. Flint, Thomas P. Divine Providence: The Molinist Account. Ithaca, NY: Cornell University Press, 1998. Frame, John M. No Other God: A Response to Open Theism. Phillipsburg, NJ: P&R Publishing, 2001. Freddoso, Alfred J. “God’s General Concurrence with Secondary Causes: Why Conservation Is Not Enough.” Philosophical Perspectives 5, Philosophy of Religion (1991): 553–85. Freddoso, Alfred J. “God’s General Concurrence with Secondary Causes: Pitfalls and Prospects.” American Catholic Philosophical Quarterly 67, no. 2 (1994): 131–56. Freddoso, Alfred J. “Introduction.” In Suarez, ed., On Creation, Conservation, and Concurrence, xi–cxxiii. Furlong, Peter. The Challenges of Divine Determinism: A Philosophical Analysis. Cambridge: Cambridge University Press, forthcoming. Gale, Richard, and Alexander R. Pruss. “A New Cosmological Argument.” Religious Studies 35, no. 4 (1999): 461–76. Garrigou-Lagrange, Reginald. Predestination. Rockford, IL: TAN Books, 1998. Geach, Peter. “Good and Evil.” Analysis 17, no. 2 (1956): 33–42. Geach, Peter. God and the Soul. New York: Schocken Books, Inc., 1969. Goris, Harm J. M.J. Free Creatures of an Eternal God: Thomas Aquinas on God’s Infallible Foreknowledge and Irresistible Will. Leuven: Peeters, 1996. Grant, W. Matthews. “Must a Cause Be Really Related to Its Effect? The Analogy between Divine and Libertarian Agent Causality.” Religious Studies 43, no. 1 (2007): 1–23. Grant, W. Matthews. “Aquinas on How God Causes the Act of Sin without Causing Sin Itself.” The Thomist 73, no. 3 (2009): 455–96. Grant, W. Matthews. “Can a Libertarian Hold That Our Free Acts Are Caused by God?” Faith and Philosophy 27, no. 1 (2010): 22–44. Grant, W. Matthews. “Anselm on Freedom: A Defense of Rogers’s Project, A Critique of Her Reconciliation of Libertarian Freedom with God the Creator Omnium.” The Saint Anselm Journal 8, no. 1 (2012). https://www.anselm.edu/sites/default/files/Documents/ Institute%20of%20SA%20Studies/Anselm%20on%20Freedom.pdf. Grant, W. Matthews. “Divine Simplicity, Contingent Truths, and Extrinsic Models of Divine Knowing.” Faith and Philosophy 29, no. 3 (2012): 254–74. Grant, W. Matthews. “Aseity.” In Routledge Encyclopedia of Philosophy. Taylor and Francis. 2015. http://www.rep.routledge.com/articles/thematic/aseity/v-1., doi: 10.4324/9780415249126-K3570-1. Grant, W. Matthews. “The Privation Account of Moral Evil: A Defense.” International Philosophical Quarterly 55, no. 3 (2015): 271–86. Grant, W. Matthews. “Divine Universal Causality and Libertarian Freedom.” In Free Will & Theism: Connections, Contingencies, and Concerns. Edited by Kevin Timpe and Daniel Speak. Oxford: Oxford University Press, 2016, 214–33. Grant, W. Matthews. “Divine Universal Causality without Occasionalism (and with AgentCausation).” In Free Will and Classical Theism: The Significance of Freedom in Perfect Being Theology. Edited by Hugh J. McCann. Oxford: Oxford University Press, 2017, 175–200.

236

Bibliography

Grant, W. Matthews. “Moral Evil, Privation, and God.” European Journal for Philosophy of Religion 9, no. 1 (2017): 125–45. Grant, W. Matthews., and Mark K. Spencer. “Activity, Identity, and God: A Tension in Aquinas and His Interpreters.” Studia Neoaristotelica: A Journal of Analytic Scholasticism 12, no. 2 (2015): 5–61. Grisez, Germain. God? A Philosophical Preface to Faith. South Bend, IN: St. Augustine’s Press, 2005. Hasker, William. “Yes, God Has Beliefs!” Religious Studies 24, no. 3 (1988): 385–94. Hasker, William. God, Time, and Knowledge. Ithaca, NY: Cornell University Press, 1989. Hasker, William. “Providence and Evil: Three Theories.” Religious Studies 28, no. 1 (1992): 91–105. Hasker, William. The Emergent Self. Ithaca, NY: Cornell University Press, 1999. Hasker, William. Providence, Evil, and the Openness of God. New York: Routledge, 2004. Hasker, William. The Triumph of God over Evil: Theodicy for a World of Suffering. Downers Grove, IL: InterVarsity Press, 2008. Hayes, John H. Amos: The Eighth Century Prophet. Nashville, Abingdon Press, 1988. Helm, Paul. The Providence of God. Downers Grove, IL: InterVarsity Press, 1993. Henninger, Mark G. “Aquinas on the Ontological Status of Relations.” Journal of the History of Philosophy 25, no. 4 (1987): 491–514. Henninger, Mark G. Relations: Medieval Theories 1250–1325. Oxford: Oxford University Press, 1989. Hinchliff, Mark. “A Defense of Presentism.” PhD diss., Princeton University, 1988. Hornsby, Jennifer, and Naomi Goulder. “Action.” In Routledge Encyclopedia of Philosophy. Taylor and Francis. 2011. http://www.rep.routledge.com/articles/thematic/action/v-2, doi: 10.4324/9780415249126-V001-2. Howard-Snyder, Daniel, and Adam Green. “Hiddenness of God.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Winter 2016 Edition, http:// plato.stanford.edu/archives/win2016/entries/divine-hiddenness/. Hütter, Reinhard. Dust Bound for Heaven: Explorations in the Theology of Thomas Aquinas. Grand Rapids, MI: Eerdmans, 2012. Johnson, Daniel M. “Calvinism and the Problem of Evil: A Map of the Territory.” In Calvinism and the Problem of Evil. Edited by David E. Alexander and Daniel M. Johnson. Eugene, OR: Pickwick Publications, 2016, 19–55. Judisch, Neal. “Meticulous Providence and Gratuitous Evil.” Oxford Studies in Philosophy of Religion 4 (2012): 67–85. Julian, of Norwich. Showings. Translated by Edmund Colledge and James Walsh. New York: Paulist Press, 1978. Kane, G. Stanley. “Evil and Privation.” International Journal for the Philosophy of Religion 11, no. 1 (1980): 43–58. Kane, Robert. The Significance of Free Will. Oxford: Oxford University Press, 1998. Kane, Robert. “Introduction: The Contours of Contemporary Free Will Debates.” In Robert Kane, ed., The Oxford Handbook of Free Will, 3–41. Kane, Robert, ed. The Oxford Handbook of Free Will. Oxford: Oxford University Press, 2002. Kane, Robert. A Contemporary Introduction to Free Will. Oxford: Oxford University Press, 2005. Keller, Simon. “Truth and Truthmaking.” Oxford Studies in Metaphysics 1 (2004): 83–104. Kenny, Anthony. Aquinas on Mind. London: Routledge, 1993.

Bibliography

237

Kim, Jaegwon. “Mechanism, Purpose, and Explanatory Exclusion.” Philosophical Perspectives, 3, Philosophy of Mind and Action Theory (1989): 77–108. Kim, Jaegwon. Physicalism, or Something Near Enough. Princeton: Princeton University Press, 2005. Kittle, Simon. “Grace and Free Will: Quiescence and Control.” Journal of Analytic Theology 3 (2015): 89–108. Klubertanz, George P. Introduction to the Philosophy of Being. New York: AppletonCentury-Crofts, Inc., 1955. Koons, Robert C. “A New Look at the Cosmological Argument.” American Philosophical Quarterly 34, no. 2 (1997): 193–211. Koons, Robert C. “Dual Agency: A Thomistic Account of Providence and Human Freedom.” Philosophia Christi 4, no. 2 (2002): 397–411. Kvanvig, Jonathan L., and Hugh J. McCann. “Divine Conservation and the Persistence of the World.” In Morris, ed., Divine and Human Action, 13–49. Lamont, John. “The Justice and Goodness of Hell.” Faith and Philosophy 28, no. 2 (2011): 152–73. Langford, M.J. “Predestination.” In The Westminster Dictionary of Christian Theology. Edited by Alan Richardson and John Bowden. Philadelphia: The Westminster Press, 1983. Lee, Patrick. “The Goodness of Creation, Evil, and Christian Teaching.” The Thomist 64, no. 2 (2000): 239–69. Lee, Patrick. “Evil as Such Is a Privation: A Reply to John Crosby.” American Catholic Philosophical Quarterly 81, no. 3 (2007): 469–88. Leftow, Brian. “Is God an Abstract Object?” Nous 24, no. 4 (1990): 581–98. Leftow, Brian. Time and Eternity. Ithaca, NY: Cornell University Press, 1991. Leftow, Brian. “Simplicity, Divine.” In Routledge Encyclopedia of Philosophy. Taylor and Francis. 1998. http://www.rep.routledge.com/articles/thematic/simplicity-divine/v-1., doi: 10.4324/9780415249126-K094-1. Leftow, Brian. “Anselm’s Perfect Being Theology.” In The Cambridge Companion to Anselm. Edited by Brian Davies and Brian Leftow. Cambridge: Cambridge University Press, 2004, 132–56. Leftow, Brian. God and Necessity. Oxford: Oxford University Press, 2012. Leibniz, Gottfried Wilhelm. Confessio Philosophi: Papers concerning the Problem of Evil, 1671–1678. Edited and Translated by Robert C. Sleigh, Jr. New Haven, CT: Yale University Press, 2005. Levering, Matthew. Predestination: Biblical and Theological Paths. Oxford: Oxford University Press, 2011. Lewis, Charles. “Divine Goodness and Worship Worthiness.” International Journal for Philosophy of Religion 14, no. 3 (1983): 143–58. Liturgy of the Hours, Vols. 1–4. New York: Catholic Book Publishing Co., 1975. Lonergan, Bernard J.F. Grace and Freedom: Operative Grace in the Thought of St. Thomas Aquinas. New York: Herder and Herder, 1971. Long, Steven A. “St. Thomas Aquinas, Divine Causality, and the Mystery of Predestination.” In Nutt, Long, and White, eds., Thomism & Predestination, 51–76. Long, Steven A., Roger W. Nutt, and Thomas Joseph White, eds. Thomism & Predestination: Principles and Disputations. Ave Maria, FL: Sapientia Press, 2016. Lowe, E.J. The Possibility of Metaphysics. Oxford: Oxford University Press, 1998. Lowe, E.J. A Survey of Metaphysics. Oxford: Oxford University Press, 2002. Lowe, E.J. Personal Agency. Oxford: Oxford University Press, 2008.

238

Bibliography

Macdonald Jr., Paul A. “Hell, the Problem of Evil, and the Perfection of the Universe.” American Catholic Philosophical Quarterly 89, no. 4 (2015): 603–28. Maimonides. Mishneh Torah. In A Maimonides Reader. Edited by Isadore Twersky. New York: Behrman House, Inc., 1972. Malebranche, Nicholas. Dialogues on Metaphysics and on Religion. Edited by Nicholas Jolley and David Scott. New York: Cambridge University Press, 1997. Marshall, Dan, and Brian Weatherson. “Intrinsic vs. Extrinsic Properties.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Spring 2018 Edition, https:// plato.stanford.edu/archives/spr2018/entries/intrinsic-extrinsic/. Matava, R.J. Divine Causality and Human Free Choice: Domingo Báñez, Physical Premotion and the Controversy de Auxiliis Revisited. Leiden: Brill, 2016. McCabe, Herbert. God Matters. Springfield, IL: Templegate Publishers, 1987. McCann, Hugh J. “Divine Sovereignty and the Freedom of the Will.” Faith and Philosophy 12, no. 4 (1995): 582–98. McCann, Hugh J. Creation and the Sovereignty of God. Bloomington: Indiana University Press, 2012. McCann, Hugh J., and Daniel M. Johnson. “Divine Providence.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Spring 2017 Edition, http:// plato.stanford.edu/archives/spr2017/entries/providence-divine/. McCann, Hugh J., and Jonathan L. Kvanvig. “The Occasionalist Proselytizer: A Modified Catechism.” Philosophical Perspectives 5, Philosophy of Religion (1991): 587–615. Merricks, Trenton. Truth and Ontology. Oxford: Oxford University Press, 2007. Miller, Barry. A Most Unlikely God: A Philosophical Enquiry. Notre Dame: University of Notre Dame Press, 1996. Miller, Barry. The Fullness of Being: A New Paradigm for Existence. Notre Dame: University of Notre Dame Press, 2002. Miller, Timothy D. “Continuous Creation and Secondary Causes: The Threat of Occasionalism.” Religious Studies 47, no. 1 (2011): 3–22. Molnar, George. Powers: A Study in Metaphysics. Oxford: Oxford University Press, 2003. Morris, Thomas V. Anselmian Explorations: Essays in Philosophical Theology. Notre Dame: University of Notre Dame Press, 1987. Morris, Thomas V. Our Idea of God: An Introduction to Philosophical Theology. Notre Dame: University of Notre Dame Press, 1991. Morris, Thomas V. “Perfection and Creation.” In Reasoned Faith: Essays in Philosophical Theology in Honor of Norman Kretzmann. Edited by Eleonore Stump. Ithaca, NY: Cornell University Press, 1993, 234–47. Morris, Thomas V., ed. Divine and Human Action: Essays in the Metaphysics of Theism. Ithaca, NY: Cornell University Press, 1988. Moser, Paul K. “Prayer.” Website entry at http://pmoser.sites.luc.edu/idolanon/Prayer.html. Accessed June 6, 2018. Murphy, Mark C. God’s Own Ethics: Norms of Divine Agency and the Argument from Evil. Oxford: Oxford University Press, 2017. Nash, Scott. “Heart.” In Mercer Dictionary of the Bible. Edited by Watson E. Mills and others. Macon, GA: Mercer University Press, 1990, 360. Newlands, Samuel. “Leibniz on Privations, Limitations, and the Metaphysics of Evil.” Journal of the History of Philosophy 52, no. 2 (2014): 281–308. Noonan, Harold. Personal Identity. New York: Routledge, 1989. O’Connor, Timothy. “Agent Causation.” In Agents, Causes, and Events. Edited by Timothy O’Connor. Oxford: Oxford University Press, 1995, 173–200.

Bibliography

239

O’Connor, Timothy. “Simplicity and Creation.” Faith and Philosophy 16, no 3. (1999): 405–12. O’Connor, Timothy. Persons and Causes. Oxford: Oxford University Press, 2000. O’Connor, Timothy. Theism and Ultimate Explanation: The Necessary Shape of Contingency. Oxford: Blackwell Publishing, 2008. O’Connor, Timothy. “Free Will.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Summer 2016 Edition, https://plato.stanford.edu/archives/sum2016/ entries/freewill. Oderberg, David S. Real Essentialism. New York: Routledge, 2007. Oderberg, David S. “Divine Premotion.” International Journal for Philosophy of Religion 79, no. 3 (2016): 207–22. Osborne Jr., Thomas M. “Divine Providence: Thomist Premotion and Contemporary Philosophy of Religion.” Nova et Vetera 4 (2006): 607–32. Ott, Ludwig. Fundamentals of Catholic Dogma. Charlotte: Tan Books, 2009. Owens, Joseph. An Elementary Christian Metaphysics. Houston: Center for Thomistic Studies, 1985. Pawl, Timothy. “Free Will and Grace.” In The Routledge Companion to Free Will. Edited by Kevin Timpe, Meghan Griffith, and Neil Levy. New York: Routledge, 2017. Pawl, Timothy, and Kevin Timpe. “Incompatibilism, Sin, and Free Will in Heaven.” Faith and Philosophy 26, no. 4 (2009): 398–419. Pereboom, Derk. Living without Free Will. Cambridge: Cambridge University Press, 2001. Pereboom, Derk. “Libertarianism and Theological Determinism.” In Timpe and Speak, eds., Free Will & Theism, 112–31. Peterson, David. G. The Acts of the Apostles. Grand Rapids, MI: Eerdmans, 2009. Pinnock, Clark, and Richard Rice, John Sanders, William Hasker, and David Basinger. The Openness of God. Downers Grove, IL: InterVarsity Press, 1994. Pinnock, Clark H. “Systematic Theology.” In Pinnock and others, eds., The Openness of God, 101–25. Piper, John. “Are There Two Wills in God?” In Still Sovereign: Contemporary Perspectives on Election, Foreknowledge, and Grace. Edited by Thomas R. Schreiner and Bruce A. Ware. Grand Rapids, MI: Baker Books, 2004, 107–31. Plantinga, Alvin. The Nature of Necessity. Oxford: Oxford University Press, 1974. Plantinga, Alvin. Does God Have a Nature? Milwaukee, WI: Marquette University Press, 1980. Plantinga, Alvin. “Supralapsarianism, or ‘O Felix Culpa.’” In Christian Faith and the Problem of Evil. Edited by Peter van Inwagen. Grand Rapids, MI: Eerdmans, 2004, 1–25. Pohle, Joseph. “Predestination.” In The Catholic Encyclopedia. New York: Robert Appleton Company, 1911. Retrieved May 24, 2017 from New Advent http://www.newadvent.org/ cathen/1237a.htm. Prestige, G.L. God in Patristic Thought. London: S.P.C.K., 1959. Pruss, Alexander R. “On Two Problems of Divine Simplicity.” Oxford Studies in the Philosophy of Religion 1 (2008): 150–67. Quinn, Philip L. “Divine Conservation, Secondary Causes, and Occasionalism.” In Morris, ed., Divine and Human Action, 50–73. Ragland, C.P. “The Trouble with Quiescence: Stump on Grace and Freedom.” Philosophia Christi 8, no. 2 (2006): 343–62. Reichberg, Gregory M. “Beyond Privation: Moral Evil in Aquinas’s De Malo.” The Review of Metaphysics 55, no. 4 (2002): 751–84.

240

Bibliography

Rhoda, Alan R. “Gratuitous Evil and Divine Providence.” Religious Studies 46, no. 3 (2010): 281–302. Rice, Richard. “Biblical Support for a New Perspective.” In Pinnock and others, eds., The Openness of God, 11–58. Rogers, Katherin A. Perfect Being Theology. Edinburgh: Edinburgh University Press, 2000. Rogers, Katherin A. “What’s Wrong with Occasionalism?” American Catholic Philosophical Quarterly 75, no. 3 (2001): 345–69. Rogers, Katherin A. “Does God Cause Sin? Anselm of Canterbury versus Jonathan Edwards on Human Freedom and Divine Sovereignty.” Faith and Philosophy 20, no. 3 (2003): 371–78. Rogers, Katherin A. “God Is Not the Author of Sin: An Anselmian Response to McCann.” Faith and Philosophy 24, no. 3 (2007): 300–10. Rogers, Katherin A. Anselm on Freedom. Oxford: Oxford University Press, 2008. Rogers, Katherin A. “Anselm against McCann on God and Sin.” Faith and Philosophy 28, no. 4 (2011): 397–415. Rogers, Katherin A. “Anselmian Theism and Created Freedom: Response to Grant and Staley.” The Saint Anselm Journal 8, no. 1 (2012). https://www.anselm.edu/sites/default/ files/Documents/Institute%20of%20SA%20Studies/Anselmian%20Theism%20and%20 Created%20Freedom.pdf. Rogers, Katherin A. Freedom and Self-Creation: Anselmian Libertarianism. Oxford: Oxford University Press, 2015. Ross, Allen P. Creation & Blessing: A Guide to the Study and Exposition of Genesis. Grand Rapids, MI: Baker Books, 1996. Ross, James F. “Creation.” The Journal of Philosophy 77, no. 10 (1980): 614–24. Ross, James F. “Creation II.” In The Existence and Nature of God. Edited by Alfred J. Freddoso. Notre Dame: University of Notre Dame Press, 1983, 115–41. Ross, James F. “Aquinas’s Exemplarism, Aquinas’s Voluntarism.” American Catholic Philosophical Quarterly 64, no. 2 (1990): 171–98. Rota, Michael. “The Eternity Solution to the Problem of Human Freedom and Divine Foreknowledge.” European Journal for Philosophy of Religion 2, no. 1 (2010): 165–86. Rota, Michael. “Causation.” In The Oxford Handbook of Aquinas. Edited by Brian Davies and Eleonore Stump. Oxford: Oxford University Press, 2012, 104–14. Sanders, John. The God Who Risks. 2nd ed. Downers Grove, IL: InterVarsity Press, 2007. Searle, John. Mind: A Brief Introduction. Oxford: Oxford University Press, 2004. Shanley, Brian J. “Divine Causation and Human Freedom in Aquinas.” American Catholic Philosophical Quarterly 72, no. 1 (1998): 99–122. Shanley, Brian J. “Aquinas on God’s Causal Knowledge: A Reply to Stump and Kretzmann.” American Catholic Philosophical Quarterly 72, no. 3 (1998). Sider, Theodore. “What’s So Bad about Overdetermination?” Philosophy and Phenomenological Research 67, no. 3 (2003): 719–26. Spencer, Mark K. “Divine Causality and Creaturely Freedom: A Thomistic Personalist View.” Nova et Vetera 14, no. 3 (2016): 375–419. Stoneham, Tom. “A Neglected Account of Perception.” Dialectica 62, no. 3 (2008): 307–22. Stump, Eleonore. “Libertarian Freedom and the Principle of Alternative Possibilities.” In Faith, Freedom, and Rationality: Philosophy of Religion Today. Edited by Jeff Jordan and Daniel Howard-Snyder. Lanham, MD: Rowman & Littlefield, 1996, 73–88. Stump, Eleonore. Aquinas. London: Routledge, 2003. Stump, Eleonore, and Norman Kretzmann. “Eternity.” The Journal of Philosophy 78, no. 8 (1981): 429–58.

Bibliography

241

Stump, Eleonore, and Norman Kretzmann. “Eternity and God’s Knowledge: A Reply to Shanley.” American Catholic Philosophical Quarterly 72, no. 3 (1998): 439–45. Suarez, Francisco. On Creation, Conservation, and Concurrence: Metaphysical Disputations 20–22. Translation and introduction by Alfred J. Freddoso. South Bend, IN: St. Augustine’s Press, 2002. Sweeney, Leo. Authentic Metaphysics in an Age of Unreality. 2nd ed. New York: Peter Lang, 1993. Swinburne, Richard. The Coherence of Theism. Oxford: Oxford University Press, 1993. Swinburne, Richard. Is There a God? Oxford: Oxford University Press, 1996. Swinburne, Richard. Providence and the Problem of Evil. Oxford: Oxford University Press, 1998. Tanner, Kathryn. God and Creation in Christian Theology: Tyranny or Empowerment? Minneapolis: Fortress Press, 2005. Tanner, Norman, ed. Decrees of the Ecumenical Councils, Volume II: Trent to Vatican II. London: Sheed & Ward, 1990. Taylor, Richard. Action and Purpose. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1966. Teresa, of Avila. “In the Hands of God.” Translated by Kieran Kavanaugh, and Otilio Rodriguez. In The Collected Works of St. Teresa of Avila, Vol. 3. Washington, DC: Institute of Carmelite Studies, 1985. Te Velde, Rudi. Aquinas on God: The “Divine Science” of the Summa Theologiae. Burlington, VT: Ashgate, 2006. Thau, Michael. Consciousness and Cognition. Oxford: Oxford University Press, 2002. Thau, Michael. “Precis.” Philosophical Studies 132, no. 3 (2007): 565–70. Timpe, Kevin. “Grace and Controlling What We Do Not Cause.” Faith and Philosophy 24, no. 3 (2007): 284–99. Timpe, Kevin. Free Will: Sourcehood and Its Alternatives. 2nd ed. London: Bloomsbury Academic, 2013. Timpe, Kevin, and Daniel Speak, eds. Free Will & Theism: Connections, Contingencies, and Concerns. Oxford: Oxford University Press, 2016. Trabbic, Joseph G. “Praemotio Physica and Divine Transcendence.” In Nutt, Long, and White, eds., Thomism & Predestination, 152–65. Turner, Denys. Thomas Aquinas: A Portrait. New Haven, CT: Yale University Press, 2013. Turner, Jason. “Compatibilism and the Free Will Defense.” Faith and Philosophy 30, no. 2 (2013): 125–37. Vallicella, William F. “Concurrentism or Occasionalism?” American Catholic Philosophical Quarterly 70, no. 3 (1996): 339–59. Vallicella, William F. A Paradigm Theory of Existence: Onto-Theology Vindicated. Dordrecht: Kluwer Academic Publishers, 2010. van Inwagen, Peter. An Essay on Free Will. Oxford: Oxford University Press, 1983. van Inwagen, Peter. “The Place of Chance in a World Sustained by God.” In Morris, ed., Divine and Human Action, 211–35. van Inwagen, Peter. Metaphysics. Boulder, CO: Westview Press, 1993. van Inwagen, Peter. The Problem of Evil. Oxford: Oxford University Press, 2006. van Inwagen, Peter. “God and Other Uncreated Things.” In God and Metaphysics: Essays in Honor of Eleonore Stump. Edited by Kevin Timpe. New York: Routledge, 2009, 3–20.

242

Bibliography

Walker, Larry L. “Heart.” In Eerdmans Dictionary of the Bible. Edited by David Noel Freedman and others. Grand Rapids, MI: Eerdmans, 2000, 563. Ward, T.M. “Relations without Forms: Some Consequences of Aquinas’s Metaphysics of Relations.” Vivarium 48, no. 3 (2010): 279–301. Ware, Bruce A. God’s Lesser Glory: The Diminished God of Open Theism. Wheaton, IL: Crossway Books, 2000. Ware, Bruce A. Their God Is Too Small: Open Theism and the Undermining of Confidence in God. Wheaton, IL: Crossway Books, 2003. Williams, Bernard. Introduction to Ethics. New York: Harper and Row, 1972. Williamson, Timothy. Knowledge and Its Limits. Oxford: Oxford University Press, 2000. Zagzebski, Linda. “Foreknowledge and Free Will.” In The Stanford Encyclopedia of Philosophy. Edited by Edward N. Zalta. Summer 2017 Edition, https://plato.stanford. edu/archives/sum2017/entries/free-will-foreknowledge/.

Index Note: Locators with letter ‘n’ refer to notes. ability to do otherwise 7, 12, 65–7, 95, 204 n.35, 204 n.39, 204 n.45, 212 n.43 accepting the offer of salvation 172–3. See also responsibility with respect to our own salvation accidental properties 83–7, 92–3, 208 n.22 act context 81 Adams, Marilyn McCord 218 n.15 Adams, Robert Merrihew 224 n.33 Aertsen, Jan 193–4 n.73 agency model See extrinsic models of divine knowing agent-cause 11, 37, 48–51, 62–3, 90–2, 95, 211 n.35 Alexander, David E. 214 n.13, 215 nn.23–4 Alston, William 1, 26, 45, 147, 150, 166–71, 201 n.9 Anglin, Bill 109–10, 214 n.13 annihilate 27 Anscombe, Elizabeth 96, 185 n.48, 198 n.44, 223 n.11 Anselm 1–4, 13, 15, 22–3, 99–101, 109, 136, 184 n.39, 190 n.39, 221 n.46 antecedent, concurrent, and consequent divine action 137, 161, 167, 175, 221 n.49 Aquinas 1–5, 9, 11–13, 15, 29–33, 35–6, 38, 40, 45, 48, 57, 71–3, 79–80, 93–4, 99–102, 109, 116, 136, 157, 170, 178, 194 nn.78–9, 195 n.1, 201 nn.12–13, 202 n.17, 202 n.20, 206 nn.54–5, 206 n.4, 207 n.16, 208 n.20, 208 n.22, 209 n.23, 209 n.26–7, 210 n.28, 210–11 n.32, 211 n.38, 214 n.22, 215 n.31, 216 n.40, 217 n.13, 221 n.46, 222 n.1, 225 n.35, 226 n.53 Aristotle 58, 80, 86, 94, 101–2, 207 n.11, 214 n.20, 216 n.39

aseity See God’s aseity assistance objection to NODUC 42–4 Augustine 1, 136, 159–61, 164–5, 170, 178, 221 n.46, 225 n.41, 226 n.49, 226 n.53 Baker, Lynne Rudder 187 n.65 Báñez, Domingo 204–5 n.51 Barclay, John M.G. 21–2, 190 n.33 Barrett, C.K. 221 n.42 Barth, Karl 170 basic action 3, 61–3, 183 n.28, 203 nn.28– 9, 203 n.31–2, 211 n.33, 211 n.35 Belgic Confession 213 n.3 Bennett, Karen 198 n.43 Bergmann, Michael 1, 201 n.8 Bontly, Thomas D. 198 n.43 Boyle, Joseph 187 n.67, 204–5 n.51 Brock, Stephen L. 196 n.14 Brower, Jeffrey 1, 9, 56, 200 n.2, 201 n.8, 201 n.11, 209 nn.24–6, 212 n.42 Burrell, David B. 204–5 n.51 Calder, Todd 108–10, 214 n.14 Cambridge Change 54, 83–5, 209 n.25 Campbell, Joseph 222 n.3 Carey, Brandon 198 n.43 Catechism of the Catholic Church 171 Catholic Mass 172 causal power 37–9. See also power causing by non-performance 115–17, 215 n.38, 216 n.39 Chisholm, Roderick 8–9, 12, 50 Church Fathers 1 Clarke, Randolph 211 n.34, 212 n.44 Clarke, W. Norris 193–4 n.73, 194 n.75, 208–9 n.22 Clifford, Richard J. 221 n.43 compatibilism 5–6, 9

244

Index

complete cause 40 composition 30–2, 83–7, 208 n.20, 208 n.22 concurrence 4–5, 25–9, 38–9, 184 n.37, 192 n.55, 211 n.38, 221 n.53 concurrent necessary conditions 63–4, 69–70, 122, 203 n.33 Consequence Argument 186 n.57 contingent and necessary being 191 n.46 control See responsibility cooperate vs. co-operate 39 Copan, Paul 182 n.6 cosmological arguments from contingency 24–5 Council of Trent 100, 226 n.50 counterfactual power 69–70, 152–5, 166–7, 169 Craig, William Lane 23, 93–4, 182 n.6, 189 n.23, 191 n.42, 207 n.10, 223 n.6 Crane, Tim 222 n.3 Cranfield, C.E.B. 221 n.42 creation ex nihilo 184 n.29 Crisp, Oliver 172, 178, 230 n.94 Crisp, Thomas M. 223 n.18 Crosby, John 110–11, 214 n.14 Cross, Richard 226–7 n.54 Curley, Edwin 187 n.65 Cyr, Taylor W. 226 n.54 Davies, Brian 4, 71, 187 n.67, 207 n.12, 217 n.10 decision set 81 Dentan, R.C. 188–9 n.18 Descartes, René 213 n.7 determinism 5–8, 10–12, 60–4, 95–7 divine conservation 2–4, 16, 25–9, 38 divine freedom 56, 58, 76–9, 201 n.13, 206 n.4 divine hiddenness 132 divine–human interaction and dialogue 166–71 divine immutability 23, 92–3, 167–8, 206 n.4, 208 n.22 divine simplicity 56–8, 76–9, 92–3, 145– 146, 167–8, 201 n.12, 206 n.4, 207 n.12, 208 n.20, 208 n.22, 222 n.1 divine transcendence 12, 71–2, 206 n.54 divine universal causality (DUC) and cosmological arguments 24–5, 191 n.48

and creaturely causes 4–5, 11, 35–51 as direct 3–5, 16, 38, 192 n.62 and divine conservation 25–9 explanation and definition 1–5 implications for creaturely action 4 and participation 29–33, 192 n.67, 193 n.72, 194 n.79 and perfect being method 22–4, 190 n.39, 191 nn.41–2 and scripture 1, 15–22, 189 nn.19–20, 190 n.33 and traditional theism 1–5, 33, 184 n.37 Dodds, Michael J. 204–5 n.51 Dual Sources, introductory and summary characterization 1, 10–14, 70–3, 181, 187 n.68 Dunn, James D.G. 218 n.14, 221 n.42 Edwards, Douglas R. 188 n.18 efficient cause 37, 41–2, 48, 196 n.14. See also power Eichrodt, Walther 16–17, 21, 222 n.57 election, doctrine of See predestination entity, defined 184 n.29 eternity See God’s eternity and relation to time exemplary cause 153, 224 n.24 extrinsic model of divine agency (EM) 12, 56–60, 201 n.11, 202 n.16, 202 n.20, 202 n.22, 204 n.34 and Aquinas 79–80, 207 n.12, 207 n.15, 207 n.16 and the causal relation 87–92, 202 n.23, 203 n.27, 207 n.15 and creaturely action 60–73, 203 n.33 and determinism 60–4, 72–3, 206 n.54 and divine simplicity 56–7, 76–9 EM argued from DUC 80–7 EM compatible with DUC 87–92. and God’s action in time 167–8 and the Intelligibility Objection 92–7 and scholastic theology 75–80. See also antecedent, concurrent, and consequent divine action extrinsic models of divine agency other than EM 76, 81–2 extrinsic models of divine knowing 145–50 agency model 148–55, 223 n.11, 224 n.25

Index belief model 146–7, 222 n.3, 223 n.5 immediate cognition model 147–8, 223 n.8 Feser, Edward 195 n.85, 196 n.14, 217–18 n.13 final cause 103–5, 214 n.20 Fitzmeyer, Joseph A. 170 Flint, Thomas P. 185–6 n.49, 186 n.57, 220 n.35, 224 n.32 Flummer, Matthew T. 226 n.54 foreknowledge 13, 19–21, 151–5, 175–6, 189 nn.23–6, 224 n.25 foreknowledge dilemma See foreknowledge Frankfurt scenarios 212 n.43 Franklin, Christopher 185 n.45 Freddoso, Alfred J. 39, 80, 184 n.43, 192 n.55, 192 n.65, 196 n.14, 197 n.27, 211 n.38 Free Will Defense (FWD)13, 119–24, 216 nn.3–4, 217 n.7, 217 n.9 Furlong, Peter 204–5 n.51 FWD Assumption See Free Will Defense Gale, Richard 24, 191 n.48, 192 n.52 Garrigou-Lagrange, Reginald 185 n.48, 204–5 n.51 Geach, Peter 106, 200 n.4, 215 n.23 God’s aseity 154–5 God’s causal acts See extrinsic model of divine agency (EM) God’s eternity and relation to time 151–5, 167–8, 206 n.4 God’s goodness 124, 141–4, 176–81, 217 n.10, 222 n.58 God’s justice 142–4, 177–81 God’s knowledge 81, 222 n.1. See also extrinsic models of divine knowing; foreknowledge God’s love 143–4, 177–81 God’s reasons 58–60 God’s salvific will 177–81 God’s willing 58, 60, 73 antecedent, concurrent, and consequent divine willing 137–8 and sin 134–44 Goetz, Stewart 109–10, 214 n.13 Goris, Harm J.M.J. 204–5 n.51

245

Goulder, Naomi 183 n.28, 203 n.28 grace 14, 132, 158–66, 225 n.36–7, 225 n.41, 226 nn.49–50, 226 nn.53–4, 227 nn.55–6, 227–8 n.58 antecedent, concurrent, and consequent grace 161–5, 226 nn.51–2, 227–8 n.58 irresistible grace 164–5 Green, Adam 219 n.31 Grisez, Germain 25, 187 n.67, 204–5 n.51 Hasker, William 13, 129, 132, 134–6, 150, 196 n.14, 198 n.44, 211 n.34, 218 n.18, 219 nn.19–20, 219 nn.23–5, 224 n.31 Hayes, John H. 141 heart 17, 188 n.18 Heaven, Hell, Purgatory 176–81 Helm, Paul 218 n.14 Henninger, Mark G. 57, 209 n.26 Hinchliff, Mark 223 n.18 Hornsby, Jennifer 183 n.28, 203 n.28 Howard-Snyder, Daniel 219 n.31 Hütter, Reinhard 204–5 n.51 immediate See divine universal causality (DUC), as direct incompatibilism 5–6, 10 Independence from God defense 217 n.9 infinite regress 87–92, 96, 210 nn.31–2, 211 n.35 instrumental causes 38, 183 n.28 Intelligibility Objection to EM 92–7 intrinsic–extrinsic distinction 53–4, 200 n.6 intrinsic models of divine agency 56, 76, 82–3 scholastic intrinsic model (SIM) 77–8, 201 n.14, 206 n.6 Johnson, Daniel M. 218 n.16, 220 n.33, 220 n.35 Judisch, Neal 219 n.21 Julian, of Norwich 170–1 Kane, Robert 5–6, 68–9, 185 n.45, 185 n.49, 186 nn.51–2, 204 n.36, 204 n.44, 204 n.49, 212 nn.43–4 Kane, Stanley 109–10, 214 n.14

246

Index

Keller, Simon 223 n.18 Kenny, Anthony 187 n.65 Kim, Jaegwon 47, 198 n.43, 199 n.49, 199 n.51 Kittle, Simon 226 n.54 Klubertanz, George P. 208–9 n.22 Koons, Robert 24, 187 n.67, 204–5 n.51 Kretzmann, Norman 9, 224 n.19 Kvanvig, Jonathan 3, 36–7, 41, 48, 196 n.10, 197–8 n.32 Lamont, John 178 Langford, M.J. 229 n.80 Lee, Patrick 214 n.13, 215 n.24 Leftow, Brian 1, 190 n.36, 191 n.42, 201 n.12, 224 n.19 Leibniz 102, 112 levels of causality 197 n.29 Levering, Matthew 229 n.83 Lewis, Charles 217 n.10 libertarian freedom apparent conflict with DUC 8–10, 54–6, 187 n.65, 201 nn.8–9, 204 n.35 explanation and definition 5–10, 65 and Free Will Defense 119–24 strict and broad accounts 9–10, 54–5, 216–17 n.6, 225 n.37, 226 n.52 Liturgy of the Hours 171–2, 229 n.85 logically sufficient 6, 185 n.48 logical problem of evil 218 n.17 Lonergan, Bernard J.F. 204–5 n.51 Long, Stephen A. 104–5 n.51, 230 n.97 Lowe, E.J. 48, 196 n.14, 200–1 n.6, 203 nn.28–9, 203 n.32 Macdonald, Paul 178, 230 n.93 Maimonides 1 Makes No Difference objection to NODUC 44–5 Malebranche 36 Marshall, Dan 200 n.1 Martens, John 189 n.20 Matava, R.J. 204–5 n.51 McCabe, Herbert 25, 169–70, 204–5 n.51 McCann, Hugh J. 1, 3, 36–7, 41, 48, 148, 187 n.67, 195 n.1, 195 n.9, 196 n.10, 197–8 n.32, 201 n.8, 204–5 n.51, 220 n.33

merit See grace Merricks, Trenton 223 n.18 middle knowledge 13, 156–8, 173–4, 218 n.17 grounding objection to 156, 158, 224 n.33 Miller, Barry 148, 195 n.82, 201 n.11, 204–5 n.51 Miller, Timothy 36, 42–44 Molinism and grace 227 n.56 and predestination 173–4, 229 n.90, 230 n.91 and the problem of moral evil 13, 126–8, 216 n.3, 218 n.17, 220 n.37, 221 n.53 and providence 156–8, 188 n.70 and sin and God’s will 134–6 Molnar, George 214 n.20 moral evil See privation account of moral evil; privation solution; problem of moral evil; scripture; sin Morris, Thomas 1, 3, 23, 26, 190 n.37, 191 n.41, 191 n.44, 192 n.66, 218 n.18, 223 n.7 Moser, Paul K. 170 Murphy, Mark C. 222 n.58 Murphy, Roland E. 221 n.43 Nash, Scott 188 n.18 Newlands, Samuel 102, 214 n.15 Nicene Creed 1 Non-Occasionalist DUC (NODUC) 35–51 epistemic objection to 45–8 metaphysical objection to 41–5 Noonan, Harold 200–1 n.6 object essentialism 76–80, 82, 206 n.7 occasionalism 4, 11, 35–51, 195 n.1, 195 n.9 O’Connor, Timothy 24, 48–51, 56–8, 91, 185 n.45, 185 n.49, 192 n.52, 199 n.61, 200 n.70, 201 n.11, 203 n.29, 203 n.31, 204 n.36, 211 n.33, 212 n.42, 223 n.12 Oderberg, David S. 104–5 n.51, 195 n.82, 196 n.14 omnipotence 158 Open Theism and foreknowledge 189 n.26, 190 n.28

Index Less Risky vs. More Risky Open Theism 130 and the problem of moral evil 128–44, 219 n.20, 219 n.25 and providence 132, 155–6, 219 n.25, 220 n.33, 220 n.35 and sin and God’s will 134–6, 220 n.41 Standard Open Theism 129–30, 219 n.23 Osborne, Thomas M., Jr. 204–5 n.51 Ott, Ludwig 182 n.7, 225 nn.35–6, 227–8 n.58, 229 n.90 overdetermination 198 n.43, 199 n.51 Owen, Joseph 209 n.25 parsimony 45–8 participation 29–33, 89, 193–4 n.73, 194 n.78 Paul the Apostle, St. 126, 136, 138, 159–60, 170, 179, 218 n.14, 226 n.49 Pawl, Timothy 186 n.53, 225 nn.36–7 Pelagianism 159–60, 163, 170 Pereboom, Derk 186 n.51, 188 n.70, 224 n.31, 224 n.33 perfect being methodology 22–4 permission of sin 134–40, 142–4, 222 n.57 antecedent permissive decree 137 permissive willing See permission of sin Peterson, David G. 221 n.44 Pinnock, Clark 155–6 Piper, John 141, 221 n.45 Plantinga, Alvin 2, 3, 8, 26, 201 n.8, 218 nn.14–15 Platonism 147, 203 n.27 Pohle, Joseph 229 n.81 popular model of divine agency (PM) 55–8, 72, 75–6, 201 n.10 possible worlds 202 n.26 Powell, Mark Allan 188 n.18 power, action as exercise of 103–5, 214 n.19 prayer as God’s gift 169–71, 228–9 n.79 prayers for salvation 172, 229 nn.84–6 predestination 171–81, 229 nn.82–3, 230 n.92 presentism 150–1, 223 n.18 Prestige, G.L. 182 n.6 presuppose 62, 64, 203 n.30 Primary Cause 38 prior 6, 60–7, 203 n.30, 204 n.36, 204 n.39, 212 n.39, 216 n.5

247

privation account of moral evil 103–12, 215 nn.26–7 privation solution 13, 99–118 problem of moral evil 13, 119–44, 217 n.13, 218 nn.14–16 providence 13, 17–18, 132, 155–8 Pruss, Alexander R. 24, 56, 191 n.48, 192 n.52, 201 n.11, 212 n.42 pure act 154 Quinn, Philip 26, 36, 192 n.62, 192 n.66, 195 n.1, 201 n.9 Ragland, C.P. 226 n.54 Reichberg, Gregory M. 215 n.31 relations medium view of relations 87–8, 211 n.38 non-reductive scholastic view of relations 88, 211 n.38 real vs. rational, and God 56–7, 76, 210 n.30 reductive scholastic view of relations 88–90, 211 n.38 responsibility 6–8, 12, 20–2, 68–71, 156–8, 172, 175, 186 n.56, 204 n.44, 204 n.49, 212 n.43 responsibility with respect to our own salvation 172, 175, 229 nn.87–8 Rhoda, Alan R. 219 n.20 Rice, Richard 224 n.27 Rogers, Katherin A. 1, 185 n.44, 190 n.37, 191 n.45, 195 n.1, 221 n.51, 223 n.7 Ross, Allen P. 221 n.43 Ross, James F. 187 n.67, 204–5 n.51, 224 n.24 Rota, Michael 196 n.14, 224 n.23 Saint Paul Daily Missal 229 n.84 saturation objection to NODUC 41–2 scholastic 1, 4–5, 11–12, 29, 35, 37, 39–40, 48, 56–8, 71, 75–80, 83–5, 88–9, 103, 187 n.68, 209 n.24, 213 n.7 scholastic intrinsic model (SIM) See intrinsic models of divine agency scripture and God’s involvement in sin 17–19, 126, 136, 141, 165–6, 178–9, 189 n.22, 213 n.10, 222 n.57

248 and God’s salvific will 179, 230 nn.96–7 and grace 159–60, 226 n.49 and predestination 172, 179–81, 229 n.83, 230 n.95 support for DUC 1, 15–22, 189 nn.19–20 Searle, John 198 n.44 secondary causes 35–40, 195 n.9, 196 n.10 Second Council of Orange 170, 226 n.50, 228 n.67 Shanley, Brian 9, 204–5 n.51, 223 n.13 Sider, Theodore 41, 196 n.12, 197 n.32 sin does God cause? 12–13, 99–118, 165–6 God’s involvement in 140–4, 213 n.10, 222 n.57 God’s permission of 134–40, 142–4 relation to God’s willing 134–44. See also privation solution; scripture skeptical theism 124, 178 sovereignty See providence Spencer, Mark K. 197 n.26, 206 n.5, 207 n.16 Stoneham, Tom 222 n.3, 223 n.8 strict dependence 224 n.26 Stump, Eleonore 9, 212 n.43, 224 n.19, 226–7 n.54, 228 n.61 Suarez, Francisco 3–5, 15, 25–7, 29, 35–6, 40, 207 n.12, 210 n.31, 211 n.38, 221 n.53 substance 37–8, 84–7, 208 n.19, 208 nn.21–2

Index Sweeney, Leo 208–9 n.22, 209 n.25 Swinburne, Richard 23, 120, 124–5, 131, 187 n.65, 219 n.26 Tanner, Kathryn 204–5 n.51 Taylor, Richard 49, 211 n.34 Teresa, of Avila 170 te Velde, Rudi 194 n.74 Thau, Michael 147 Thomistic 11, 71, 187 n.68, 204–5 n.51, 207 n.16 Timpe, Kevin 185 n.45, 186 n.50, 186 n.53, 217 n.8, 226–7 n.54 Tollefsen, Olaf 187 n.67, 204–5 n.51 Trabbic, Joseph G. 204–5 n.51 Turner, Denys 71 Turner, Jason 216 n.4 ultimate responsibility See responsibility universalism See Heaven, Hell, Purgatory Vallicella, William 36–7, 195 n.82, 196 n.10, 199 n.56 van Inwagen, Peter 26, 45, 48, 186 n.57, 195 n.1, 199 n.53, 200–2 n.6, 212 n.45, 220 n.36 Walker, Larry L. 188–9 n.18 Ward, T.M. 209 nn.26–7 Weatherson, Brian 200 n.1 Westminster Confession 213 n.3 Williams, Bernard 215 n.23 Williamson, Timothy 223 n.5 Zagzebski, Linda 152, 224 n.21